Notebooks/3 Sentiment Analysis Modeling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f16b3d4e",
   "metadata": {},
   "source": [
    "# Airline Sentiment Analysis\n",
    "\n",
    "As a business changes, so do their customer interests and sentiments. Businesses can use sentiment analysis to track customer sentiment when launching a new product or changing the price of their products. Tracking sentiment analysis can help businesses improve products and services. Twitter is a great way to gather sentiment from customers because the platform was created to communicate and stay connected through the exchange of quick, frequent messages. Using Twitter is a great way to keep track of customer sentiment about your product, service, or business on social media. It also can help detect angry customers or negative comments. A business can use this for marketing. Using Twitter for your sentiment analysis can provide valuable insights that drive business decisions. What do customers like about your business? What do customers hate or love about your products? A business can also use this for customer support. When a model can predict if a tweet is positive or negative, a bot can respond with an appropriate message to help a customer. My goal is to gather the sentiment analysis of a variety of airlines. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5d4ae07",
   "metadata": {},
   "source": [
    "## Table of Contents:\n",
    "* [Data Collection](#DataCollection)\n",
    "* [Data Organization](#DataOrganization)\n",
    "* [Data Definition](#DataDefinition)\n",
    "* [Data Cleaning](#DataCleaning)\n",
    "    * [Vectorization](#Vectorization)\n",
    "* [Model](#Model)\n",
    "    * [Naive Bayes](#nb)\n",
    "    * [Random Forest Classifier](#rf)\n",
    "    * [Logistic Regression](#lr)\n",
    "    * [N-grams](#ng)\n",
    "    * [ROC-AUC Comparisons](#roc)\n",
    "    * [Precision Recall](#pr)\n",
    "    * [Thresholding](#thresholding)\n",
    "        * [Classification Report](#cr)\n",
    "        * [Confusion Matrix](#cm)\n",
    "* [Summary](#summary) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3de85920",
   "metadata": {},
   "outputs": [],
   "source": [
    "### imports ###\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "### this will deal with punctuation ###\n",
    "import string\n",
    "import re\n",
    "\n",
    "import nltk ### using the Natural Language Toolkit ###\n",
    "from nltk.corpus import stopwords\n",
    "### Lemmatization ###\n",
    "from nltk.stem import WordNetLemmatizer\n",
    "import emoji\n",
    "from nltk.tokenize import word_tokenize\n",
    "\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import math\n",
    "from sklearn.metrics import roc_curve, auc, classification_report, balanced_accuracy_score, roc_auc_score\n",
    "from sklearn.metrics import confusion_matrix, fbeta_score, recall_score, accuracy_score, f1_score, precision_score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94a9759f",
   "metadata": {},
   "source": [
    "# 1. Data Collection <a class=\"anchor\" id=\"DataCollection\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0af8dd02",
   "metadata": {},
   "outputs": [],
   "source": [
    "### get data ###\n",
    "tweets = pd.read_csv(\"../Data/Tweets.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c034e195",
   "metadata": {},
   "source": [
    "# 2. Data Organization <a class=\"anchor\" id=\"DataOrganization\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "093c1148",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>airline_sentiment</th>\n",
       "      <th>airline_sentiment_confidence</th>\n",
       "      <th>negativereason</th>\n",
       "      <th>negativereason_confidence</th>\n",
       "      <th>airline</th>\n",
       "      <th>airline_sentiment_gold</th>\n",
       "      <th>name</th>\n",
       "      <th>negativereason_gold</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>text</th>\n",
       "      <th>tweet_coord</th>\n",
       "      <th>tweet_created</th>\n",
       "      <th>tweet_location</th>\n",
       "      <th>user_timezone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>570306133677760513</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>cairdin</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica What @dhepburn said.</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:35:52 -0800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Eastern Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>570301130888122368</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.3486</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>jnardino</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica plus you've added commercials t...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:15:59 -0800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Pacific Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>570301083672813571</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.6837</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Virgin America</td>\n",
       "      <td>NaN</td>\n",
       "      <td>yvonnalynn</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>@VirginAmerica I didn't today... Must mean I n...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-02-24 11:15:48 -0800</td>\n",
       "      <td>Lets Play</td>\n",
       "      <td>Central Time (US &amp; Canada)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id airline_sentiment  airline_sentiment_confidence  \\\n",
       "0  570306133677760513           neutral                        1.0000   \n",
       "1  570301130888122368          positive                        0.3486   \n",
       "2  570301083672813571           neutral                        0.6837   \n",
       "\n",
       "  negativereason  negativereason_confidence         airline  \\\n",
       "0            NaN                        NaN  Virgin America   \n",
       "1            NaN                        0.0  Virgin America   \n",
       "2            NaN                        NaN  Virgin America   \n",
       "\n",
       "  airline_sentiment_gold        name negativereason_gold  retweet_count  \\\n",
       "0                    NaN     cairdin                 NaN              0   \n",
       "1                    NaN    jnardino                 NaN              0   \n",
       "2                    NaN  yvonnalynn                 NaN              0   \n",
       "\n",
       "                                                text tweet_coord  \\\n",
       "0                @VirginAmerica What @dhepburn said.         NaN   \n",
       "1  @VirginAmerica plus you've added commercials t...         NaN   \n",
       "2  @VirginAmerica I didn't today... Must mean I n...         NaN   \n",
       "\n",
       "               tweet_created tweet_location               user_timezone  \n",
       "0  2015-02-24 11:35:52 -0800            NaN  Eastern Time (US & Canada)  \n",
       "1  2015-02-24 11:15:59 -0800            NaN  Pacific Time (US & Canada)  \n",
       "2  2015-02-24 11:15:48 -0800      Lets Play  Central Time (US & Canada)  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### view the heading of data ###\n",
    "tweets.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4e9520db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(14640, 15)\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 14640 entries, 0 to 14639\n",
      "Data columns (total 15 columns):\n",
      " #   Column                        Non-Null Count  Dtype  \n",
      "---  ------                        --------------  -----  \n",
      " 0   tweet_id                      14640 non-null  int64  \n",
      " 1   airline_sentiment             14640 non-null  object \n",
      " 2   airline_sentiment_confidence  14640 non-null  float64\n",
      " 3   negativereason                9178 non-null   object \n",
      " 4   negativereason_confidence     10522 non-null  float64\n",
      " 5   airline                       14640 non-null  object \n",
      " 6   airline_sentiment_gold        40 non-null     object \n",
      " 7   name                          14640 non-null  object \n",
      " 8   negativereason_gold           32 non-null     object \n",
      " 9   retweet_count                 14640 non-null  int64  \n",
      " 10  text                          14640 non-null  object \n",
      " 11  tweet_coord                   1019 non-null   object \n",
      " 12  tweet_created                 14640 non-null  object \n",
      " 13  tweet_location                9907 non-null   object \n",
      " 14  user_timezone                 9820 non-null   object \n",
      "dtypes: float64(2), int64(2), object(11)\n",
      "memory usage: 1.7+ MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "### check shape ###\n",
    "print(tweets.shape)\n",
    "print(tweets.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "30e2c5db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tweet_id                            0\n",
      "airline_sentiment                   0\n",
      "airline_sentiment_confidence        0\n",
      "negativereason                   5462\n",
      "negativereason_confidence        4118\n",
      "airline                             0\n",
      "airline_sentiment_gold          14600\n",
      "name                                0\n",
      "negativereason_gold             14608\n",
      "retweet_count                       0\n",
      "text                                0\n",
      "tweet_coord                     13621\n",
      "tweet_created                       0\n",
      "tweet_location                   4733\n",
      "user_timezone                    4820\n",
      "dtype: int64\n",
      "tweet_id                         0.000000\n",
      "airline_sentiment                0.000000\n",
      "airline_sentiment_confidence     0.000000\n",
      "negativereason                  37.308743\n",
      "negativereason_confidence       28.128415\n",
      "airline                          0.000000\n",
      "airline_sentiment_gold          99.726776\n",
      "name                             0.000000\n",
      "negativereason_gold             99.781421\n",
      "retweet_count                    0.000000\n",
      "text                             0.000000\n",
      "tweet_coord                     93.039617\n",
      "tweet_created                    0.000000\n",
      "tweet_location                  32.329235\n",
      "user_timezone                   32.923497\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "### check for null values ###\n",
    "print(tweets.isnull().sum())\n",
    "\n",
    "### percentage of null values ###\n",
    "print(((tweets.isnull().sum() * 100)/ len(tweets.index)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31c45b9d",
   "metadata": {},
   "source": [
    "\"airline_sentiment_gold\" , \"negativereason_gold\", and \"tweet_coord\" columns are missing over 90% of their input. I will drop them"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e6b319e8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tweet_id                           0\n",
      "airline_sentiment                  0\n",
      "airline_sentiment_confidence       0\n",
      "negativereason                  5462\n",
      "negativereason_confidence       4118\n",
      "airline                            0\n",
      "name                               0\n",
      "retweet_count                      0\n",
      "text                               0\n",
      "tweet_created                      0\n",
      "tweet_location                  4733\n",
      "user_timezone                   4820\n",
      "dtype: int64\n",
      "tweet_id                         0.000000\n",
      "airline_sentiment                0.000000\n",
      "airline_sentiment_confidence     0.000000\n",
      "negativereason                  37.308743\n",
      "negativereason_confidence       28.128415\n",
      "airline                          0.000000\n",
      "name                             0.000000\n",
      "retweet_count                    0.000000\n",
      "text                             0.000000\n",
      "tweet_created                    0.000000\n",
      "tweet_location                  32.329235\n",
      "user_timezone                   32.923497\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "tweets = tweets.drop(columns=[\"airline_sentiment_gold\" , \"negativereason_gold\", \"tweet_coord\"])\n",
    "\n",
    "### check for more null values ###\n",
    "print(tweets.isnull().sum())\n",
    "print(((tweets.isnull().sum() * 100)/ len(tweets.index)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b7bff738",
   "metadata": {},
   "outputs": [],
   "source": [
    "### remove tweet_id duplicates ###\n",
    "tweets = tweets.drop_duplicates(subset=[\"tweet_id\"], keep=False) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c0776df0",
   "metadata": {},
   "source": [
    "# 3. Data Definition <a class=\"anchor\" id=\"DataDefinition\"></a>\n",
    "\n",
    " Below are the main columns I am working with.\n",
    "                    \n",
    "* <b>Airline_sentiment</b>\n",
    "    * sentiment of the airline, has 3 different sentiments: positive, neutral and negative\n",
    "    * object data type\n",
    "* <b>Negativereason</b>\n",
    "    * reason for the negative sentiment \n",
    "    * Object data type\n",
    "* <b>Airline</b>\n",
    "    * name of the airline\n",
    "    * Object data type\n",
    "* <b>Text</b>\n",
    "    * The tweet\n",
    "    * Object"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e02d78c2",
   "metadata": {},
   "source": [
    "# 4. Data Cleaning <a class=\"anchor\" id=\"DataCleaning\"></a>\n",
    "\n",
    "The following needs to be done to preprocess the data:\n",
    "\n",
    "1. Make text lowercase\n",
    "2. Removing punctuations, URLs, names\n",
    "3. Tokenization\n",
    "4. Removing stopwords (\"this\", \"is\", etc.)\n",
    "4. Lemmatizing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "28ea5207",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Lowercase ###\n",
    "tweets[\"text\"] = tweets[\"text\"].str.lower()\n",
    "\n",
    "### remove numbers ###\n",
    "tweets[\"text\"] = tweets[\"text\"].str.replace('\\d+', '', regex=True)\n",
    "\n",
    "### REMOVE PUNCTUATIONS ###\n",
    "def remove_punc(text):\n",
    "    words_wo_punct = re.sub(r\"[^A-Za-z0-9\\s]+\", \"\", text)\n",
    "    return words_wo_punct\n",
    "\n",
    "tweets[\"text\"] = tweets[\"text\"].apply(lambda x: remove_punc(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "65b03e7f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                     virginamerica what dhepburn said\n",
       "1    virginamerica plus youve added commercials to ...\n",
       "2    virginamerica i didnt today must mean i need t...\n",
       "3    virginamerica its really aggressive to blast o...\n",
       "4    virginamerica and its a really big bad thing a...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### REMOVE WHITESPACE ###\n",
    "tweets[\"text\"] = tweets[\"text\"].str.strip()\n",
    "\n",
    "### remove emoji ###\n",
    "tweets[\"text\"] = tweets[\"text\"].apply(lambda x: emoji.demojize(x))\n",
    "tweets[\"text\"].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c3391dba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                     virginamerica what dhepburn said\n",
       "1    virginamerica plus youve added commercial to t...\n",
       "2    virginamerica i didnt today must mean i need t...\n",
       "3    virginamerica it really aggressive to blast ob...\n",
       "4    virginamerica and it a really big bad thing ab...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### LEMMATIZING ###\n",
    "lemmatizer = WordNetLemmatizer()\n",
    "\n",
    "def word_lemmatizer(text):\n",
    "    word_list = nltk.word_tokenize(text)\n",
    "    lem_text = ' '.join([lemmatizer.lemmatize(w) for w in word_list])\n",
    "    return lem_text\n",
    "\n",
    "tweets[\"text\"] = tweets[\"text\"].apply(lambda x: word_lemmatizer(x))\n",
    "tweets.text.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "32b92c5b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                      [virginamerica, dhepburn, said]\n",
       "1    [virginamerica, plus, youve, added, commercial...\n",
       "2    [virginamerica, didnt, today, must, mean, need...\n",
       "3    [virginamerica, really, aggressive, blast, obn...\n",
       "4             [virginamerica, really, big, bad, thing]\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### REMOVE STOPWORDS and tokenize ###\n",
    "def remove_stopwords(text):\n",
    "    tokens = word_tokenize(text)\n",
    "    english_stopwords = stopwords.words('english')\n",
    "    tokens_wo_stopwords = [t for t in tokens if t not in english_stopwords]\n",
    "    \n",
    "    return tokens_wo_stopwords\n",
    "\n",
    "tweets[\"text\"] = tweets[\"text\"].apply(lambda x: remove_stopwords(str(x)))\n",
    "tweets[\"text\"].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30186610",
   "metadata": {},
   "source": [
    "### Vectorization <a class=\"anchor\" id=\"Vectorization\"></a>\n",
    "\n",
    "Vectorize text with Logistic Regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "eb9d6cad",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer, TfidfTransformer\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3bc1db9e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14330, 12)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "89983bfc",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Change Text to String ###\n",
    "tweets[\"text\"] = tweets[\"text\"].astype(str)\n",
    "\n",
    "### Remove neutral sentiment to focus on just the positive and negative sentiment ###\n",
    "tweets = tweets[tweets[\"airline_sentiment\"] != \"neutral\"]\n",
    "\n",
    "### Change sentiment to int ###\n",
    "def changeSentiment(sentiment):\n",
    "    if  sentiment == \"positive\":\n",
    "        return 0\n",
    "    elif sentiment == \"negative\":\n",
    "        return 1\n",
    "    \n",
    "tweets['airline_sentiment'] = tweets['airline_sentiment'].apply(lambda x : changeSentiment(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "edb373a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11295, 12)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08ebcf8b",
   "metadata": {},
   "source": [
    "### CountVec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "01192ef9",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Train Test Split ###\n",
    "cv = CountVectorizer(min_df=5, max_df=0.70)\n",
    "X = cv.fit_transform(tweets.text)\n",
    "y = tweets['airline_sentiment']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.30, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5611ae14",
   "metadata": {},
   "source": [
    "#### Get base model with Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f2a1e1db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression Accuracy: 92.18%\n",
      "Accuracy on training data: 0.95\n",
      "Accuracy on test data:     0.92\n",
      "0.9577395058139224\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "logReg = LogisticRegression(max_iter=1000).fit(X_train, y_train)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Logistic Regression Accuracy: %0.2f%%\" % (100 * logReg.score(X_test, y_test)))\n",
    "\n",
    "training_accuracy = logReg.score(X_train, y_train)\n",
    "test_accuracy = logReg.score(X_test, y_test)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test, logReg.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe7e7177",
   "metadata": {},
   "source": [
    "### Tf-idf "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a4d8905b",
   "metadata": {},
   "outputs": [],
   "source": [
    "td = TfidfVectorizer(min_df=5, max_df=0.70)\n",
    "X = td.fit_transform(tweets['text'])\n",
    "y = tweets['airline_sentiment']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.30, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7bfb15e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression Accuracy: 90.94%\n",
      "Accuracy on training data: 0.92\n",
      "Accuracy on test data:     0.91\n",
      "0.9583204152939573\n"
     ]
    }
   ],
   "source": [
    "logReg2 = LogisticRegression(max_iter=1000).fit(X_train, y_train)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Logistic Regression Accuracy: %0.2f%%\" % (100 * logReg2.score(X_test, y_test)))\n",
    "\n",
    "training_accuracy = logReg2.score(X_train, y_train)\n",
    "test_accuracy = logReg2.score(X_test, y_test)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test, logReg2.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa56a743",
   "metadata": {},
   "source": [
    "Vectorizing with Tf-idf performed the best based on ROC-AUC Score."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b82ce620",
   "metadata": {},
   "source": [
    "# Model Selection <a class=\"anchor\" id=\"Model\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "04749e28",
   "metadata": {},
   "outputs": [],
   "source": [
    "tf = TfidfVectorizer(min_df=5, max_df=0.70)\n",
    "X = tf.fit_transform(tweets['text'])\n",
    "y = tweets['airline_sentiment']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.30, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "289c3acc",
   "metadata": {},
   "source": [
    "## Naive Bayes <a class=\"anchor\" id=\"nb\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "a22f7d30",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Naive Bayes Accuracy: 89.17%\n",
      "Accuracy on training data: 0.91\n",
      "Accuracy on test data:     0.89\n",
      "0.9511944850610249\n"
     ]
    }
   ],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB\n",
    "\n",
    "mnb = MultinomialNB().fit(X_train, y_train)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Naive Bayes Accuracy: %0.2f%%\" % (100 * mnb.score(X_test, y_test)))\n",
    "\n",
    "training_accuracy = mnb.score(X_train, y_train)\n",
    "test_accuracy = mnb.score(X_test, y_test)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test, mnb.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "3c369417",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Get false positive rate and the true positive rate ###\n",
    "mnb_probs = mnb.predict_proba(X_test)[:,1]\n",
    "mnb_fpr, mnb_tpr, mnb_thresholds = roc_curve(y_test, mnb_probs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a60e5dcb",
   "metadata": {},
   "source": [
    "## Random Forest Classifier <a class=\"anchor\" id=\"rf\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b4c18878",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Accuracy: 90.03%\n",
      "Accuracy on training data: 1.00\n",
      "Accuracy on test data:     0.90\n",
      "0.9548698237103328\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf = RandomForestClassifier().fit(X_train, y_train)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Random Forest Accuracy: %0.2f%%\" % (100 * rf.score(X_test, y_test)))\n",
    "\n",
    "training_accuracy = rf.score(X_train, y_train)\n",
    "test_accuracy = rf.score(X_test, y_test)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test, logReg.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb4c9ba8",
   "metadata": {},
   "source": [
    "## Lets Improve the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "76454b14",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'max_depth': None, 'max_features': 'auto', 'n_estimators': 400}\n",
      "0.9332323407683674\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "parameters = {\n",
    "    'max_depth': [None, 25, 50], \n",
    "    'max_features': ['auto'], \n",
    "    'n_estimators': [100, 300, 400]\n",
    "}\n",
    "\n",
    "rf_gs = GridSearchCV(rf, parameters, cv=5, scoring = \"roc_auc\") \n",
    "rf_gs.fit(X_train, y_train)\n",
    "\n",
    "print(rf_gs.best_params_)\n",
    "print(rf_gs.best_score_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "8de80778",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Accuracy: 89.94%\n",
      "Accuracy on training data: 1.00\n",
      "Accuracy on test data:     0.90\n",
      "0.9365687613379263\n"
     ]
    }
   ],
   "source": [
    "rf = RandomForestClassifier(max_depth= None, max_features= 'auto', n_estimators = 400).fit(X_train, y_train)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Random Forest Accuracy: %0.2f%%\" % (100 * rf.score(X_test, y_test)))\n",
    "\n",
    "training_accuracy = rf.score(X_train, y_train)\n",
    "test_accuracy = rf.score(X_test, y_test)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test, rf.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "91252c87",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Get false positive rate and the true positive rate ###\n",
    "rf_probs = rf.predict_proba(X_test)[:,1]\n",
    "rf_fpr, rf_tpr, rf_thresholds = roc_curve(y_test, rf_probs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7d39d47",
   "metadata": {},
   "source": [
    "## Logistic Regression <a class=\"anchor\" id=\"lr\"></a>\n",
    "\n",
    "The Logistic Regression model performed the best. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "9b31452d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression Accuracy: 90.94%\n",
      "Accuracy on training data: 0.92\n",
      "Accuracy on test data:     0.91\n",
      "0.9583204152939573\n"
     ]
    }
   ],
   "source": [
    "lr = LogisticRegression(max_iter=1000).fit(X_train, y_train)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Logistic Regression Accuracy: %0.2f%%\" % (100 * lr.score(X_test, y_test)))\n",
    "\n",
    "training_accuracy = lr.score(X_train, y_train)\n",
    "test_accuracy = lr.score(X_test, y_test)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test, lr.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db3ede89",
   "metadata": {},
   "source": [
    "## The base model performed the best. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6d0c823",
   "metadata": {},
   "source": [
    "## Let's improve the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "d8ad5d89",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'C': 1, 'max_iter': 1000}\n",
      "0.9499753030090758\n"
     ]
    }
   ],
   "source": [
    "### Logistic Regression GridSearchCV ###\n",
    "parameters = {\n",
    "    'C' : [.01, .001, 1, 10, 100],\n",
    "    'max_iter': [1000, 10000]\n",
    "}\n",
    "\n",
    "lr_gs = GridSearchCV(lr, parameters, cv=5, scoring = \"roc_auc\") \n",
    "lr_gs.fit(X_train, y_train)\n",
    "\n",
    "print(lr_gs.best_params_)\n",
    "print(lr_gs.best_score_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "3cb9668f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression Accuracy: 90.94%\n",
      "Accuracy on training data: 0.92\n",
      "Accuracy on test data:     0.91\n",
      "0.9583204152939573\n"
     ]
    }
   ],
   "source": [
    "lr = LogisticRegression(max_iter=1000, C=1).fit(X_train, y_train)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Logistic Regression Accuracy: %0.2f%%\" % (100 * lr.score(X_test, y_test)))\n",
    "\n",
    "training_accuracy = lr.score(X_train, y_train)\n",
    "test_accuracy = lr.score(X_test, y_test)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test, lr.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d07cbfb7",
   "metadata": {},
   "source": [
    "# Let's try with N-grams <a class=\"anchor\" id=\"ng\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea534966",
   "metadata": {},
   "source": [
    "## CountVect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "03a53792",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression Accuracy: 91.44%\n",
      "Accuracy on training data: 1.00\n",
      "Accuracy on test data:     0.91\n",
      "0.9570523912765954\n"
     ]
    }
   ],
   "source": [
    "ngram = (1,2)\n",
    "cv = CountVectorizer(ngram_range=ngram)\n",
    "X = cv.fit_transform(tweets['text'])\n",
    "y = tweets['airline_sentiment']\n",
    "X_train2, X_test2, y_train2, y_test2 = train_test_split(X, y, test_size=.30, random_state=0)\n",
    "\n",
    "lr2 = LogisticRegression(max_iter=1000, C=1).fit(X_train2, y_train2)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Logistic Regression Accuracy: %0.2f%%\" % (100 * lr2.score(X_test2, y_test2)))\n",
    "y_pred =lr2.predict(X_test2)\n",
    "training_accuracy = lr2.score(X_train2, y_train2)\n",
    "test_accuracy = lr2.score(X_test2, y_test2)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score( y_test2, lr2.predict_proba(X_test2)[:,1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83c7bbe5",
   "metadata": {},
   "source": [
    "## TF-IDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "8a9b5d38",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression Accuracy: 87.72%\n",
      "Accuracy on training data: 0.91\n",
      "Accuracy on test data:     0.88\n",
      "0.9517174645098101\n"
     ]
    }
   ],
   "source": [
    "ngram = (1,2)\n",
    "td = TfidfVectorizer(ngram_range=ngram)\n",
    "X = td.fit_transform(tweets['text'])\n",
    "y = tweets['airline_sentiment']\n",
    "X_train3, X_test3, y_train3, y_test3 = train_test_split(X, y, test_size=.30, random_state=0)\n",
    "\n",
    "lr3 = LogisticRegression(max_iter=1000, C=1).fit(X_train3, y_train3)\n",
    "\n",
    "### Print accuracy ###\n",
    "print(\"Logistic Regression Accuracy: %0.2f%%\" % (100 * lr3.score(X_test3, y_test3)))\n",
    "y_pred =lr3.predict(X_test3)\n",
    "training_accuracy = lr3.score(X_train3, y_train3)\n",
    "test_accuracy = lr3.score(X_test3, y_test3)\n",
    "\n",
    "print(\"Accuracy on training data: %0.2f\" % (training_accuracy))\n",
    "print(\"Accuracy on test data:     %0.2f\" % (test_accuracy))\n",
    "print(roc_auc_score(y_test3, lr3.predict_proba(X_test3)[:,1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "6806a7b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Get false positive rate and the true positive rate for lr ###\n",
    "lr_probs = lr.predict_proba(X_test)[:,1]\n",
    "lr_fpr, lr_tpr, lr_thresholds = roc_curve(y_test, lr_probs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c78aa03",
   "metadata": {},
   "source": [
    "### The lr performed the best based on roc_auc score and accuracy.\n",
    "\n",
    "## ROC Curve Comparisons <a class=\"anchor\" id=\"roc\"></a>\n",
    "\n",
    "Below compares the three models based on the ROC-AUC (Receiver Operating Characteristic-Area Under the Curve) to visualize the tradeoff between the classifier’s sensitivity and specificity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "34a70703",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,8))\n",
    "\n",
    "for fpr, tpr, model in [(mnb_fpr,mnb_tpr,'Naive Bayes'),(rf_fpr,rf_tpr,'Random Forest'),(lr_fpr,lr_tpr,'Logistic Regression')]:\n",
    "    roc_auc = auc(fpr, tpr)\n",
    "    plt.plot(fpr, tpr,\n",
    "             lw=2, label='ROC '+ model + ' (area = %0.2f)' % roc_auc)\n",
    "\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC-AUC comparison')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f05a8d87",
   "metadata": {},
   "source": [
    "# Precision Recall Curve <a class=\"anchor\" id=\"pr\"></a>\n",
    "\n",
    "The Precision Recall Curve shows the tradeoff between a classifier’s precision (a measure of result relevancy) and recall(a measure of completeness). For each class, precision is defined as the ratio of true positives to the sum of true and false positives, and recall is the ratio of true positives to the sum of true positives and false negatives. My model has an average precision of 0.99."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "15dacff6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA30AAAFKCAYAAACpajaFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAABFo0lEQVR4nO3dZ2AU1d7H8d/spmdD6B1CBzV0LAgRkCYIUiIG4VIUsYAVEAEFkW7gsYFgR0QFvKhUUS9dEAtIiwhBOogQpIS0TXZ3nhcLa0IChLIENt/Pi3uZnbPn/Gd3hPxyZuYYpmmaAgAAAAD4JEteFwAAAAAA8B5CHwAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4MEIfAAAAAPgwv7wuAAB8wcGDB9WyZUtVq1bN85ppmurZs6fuv//+qzLGm2++qYiICHXs2PG8bTp06KCZM2eqQIECV2XMu+++W/7+/goKCpJhGEpPT5fFYtHgwYN11113XZUxMqtevbrWrVunlStX6rvvvtO7776brY3dbte0adO0cuVKmaYpl8ul9u3bq2/fvjIM46rXdCFr1qzR8OHDVaRIEX366acKCgq65D6++uqr8x7r5cjNeTJlyhTVqFFDLVq0yFX7Hj166NChQwoLC5MkuVwupaen64knnrjg+/LKkSNH9Mwzz2j27Nl5XQoAXBcIfQBwlQQFBWn+/Pme7SNHjqhdu3aKjIxUjRo1rrj/Z5555qJtMo9/tUyaNEk1a9b0bH/77bcaNmyY1qxZc9XHuhjTNNWvXz9VrFhRc+bMUWBgoE6cOKHHHntMKSkpevbZZ69pPYsXL1aXLl3Ur1+/azruheTmPPn5559VpUqVXLeXpMGDB+uee+7xbG/dulUPPvigWrRoIZvNdnnFekmJEiUIfACQCaEPALykRIkSioiI0N69e7Vt2zbNnTtXqampstlsmjlzpv773/9q1qxZcrlcKliwoIYPH67KlSsrOTlZY8aM0W+//Sar1aoWLVroueee09ChQ1W1alX16dNHb731lv73v//J399fhQoV0vjx41W8eHHPTFnhwoX19ttva/HixbJarapYsaKGDx+uYsWKqUePHqpTp45+++03HT58WA0bNtTo0aNlsVz8in/TNHXw4EGFh4d7XrvU49i7d69GjRql5ORkJSQkqEaNGnrjjTcUGBh40fF//fVX7d69W++9956sVqskqVChQoqNjdWhQ4ckuWelunfv7gkombcjIyPVvHlzbd++Xffff782bNigd955R5K0a9cu9e7dWytXrtTevXs1duxYnTx5Uk6nUz169Mg2Y/vBBx9o2bJlCgwM1OnTpzVgwABNmDBB69atk9VqVa1atTR06FDZbDbdfffdqlWrlnbs2KEBAwaoZcuWuTqHdu7cqVGjRunkyZMyDEMPP/ywZ2btvffe09y5cxUaGqoGDRpo2bJlWr58uYYMGXLB8+R///uf4uLiFBsbK6vVqmXLlnnab968WWPGjFFqaqr8/f01ePBgNWzYMMfaDhw4oJCQEAUEBEiSli9frmnTpikjI0NBQUF64YUXVLduXaWmpurll1/W5s2bFRYW5gmbEyZMyPa51KpVS6NGjdLhw4eVkZGhe++9V48//rgcDodGjx6t3377Tf7+/ipbtqzGjx+vwMDAHF8/ceKE2rdvr40bNyojI+OC30unTp20bt06HT58WB06dLjmvzgAgGuB0AcAXrJx40bt379ftWvX1rp16/Tnn39q+fLlstls+uWXXzRv3jx99tlnCg4O1po1a/Tkk09qyZIleuutt2S32/XNN9/I6XTq4Ycf1i+//OLp9/Dhw5oxY4bWrVungIAAffTRR9qyZYtatGjhafPll1/qhx9+0Ny5cxUSEqLJkydryJAh+vDDDyVJ+/fv18yZM5WSkqI2bdrol19+0R133JHjcQwaNEiBgYE6efKkJKlx48aeoHQ5x7Fy5Up17NhRHTp0UEZGhjp37qyVK1eqdevWF/1M4+LiVKtWLU/gO6tChQqqUKHCRd+fkZGhZs2a6c0331RSUpLee+89JSQkqFixYvrqq6/UuXNnmaapp59+WrGxsbrlllt0+vRpxcTEqEqVKqpTp46nr0ceeUR//vlnloB19OhRzZ8/X1arVS+++KJiY2M1atQoSVLVqlX1xhtvXLTGsxwOh5544gkNHjxYrVq10pEjR9SlSxdFREQoKSlJX331lebOnauwsDC9+OKL2d5/vvOke/fu+vbbb9W9e3e1bNlSy5Yt83w2/fv315gxY9S0aVPFxcVp6NChntnj2NhYTZs2TYmJibLb7brjjjv08ccfKyAgQHv37tXrr7+uTz75RIUKFdLOnTv10EMP6fvvv9e0adPkdDq1ZMkSpaSkqFu3brr55ps9dWb+XHr27KnevXvr7rvvlt1uV9++fVW+fHkVL15cv/zyi7755hsZhqGJEydqx44dcrlcOb5evHhxT//Tpk274PeSkpKizz//XEeOHFHLli0VHR2tcuXK5fp7AoAbAaEPAK6StLQ0dejQQZLkdDpVqFAhTZw4UaVKlZLkvl/t7GVwK1eu1L59+9S1a1fP+xMTE3Xy5En9+OOPGjp0qKxWq6xWqz799FNJ0tdffy3JPYNYo0YNderUSXfddZfuuuuubLMxq1evVufOnRUSEiLJ/cP0O++8o/T0dElSs2bNZLFYZLPZFBERoVOnTp33uM5e3nngwAE99NBDuummmzw/FF/Ocdx6661au3at3n//fe3du1dHjx5VSkpKrj5ji8Ui0zRz1fZ8GjRoIEmy2Wxq2bKlFixYoN69e2vhwoX67LPPtHfvXu3fv1/Dhg3zvCctLU3btm3LEvrOtXr1aj333HPy9/eX5J5h7N+/f7Zxc2vv3r2y2+1q1aqVJPf33qpVK/3www9KTEzUPffc47l3s3v37vrpp5+yvD8350lm8fHxslgsatq0qSQpMjJSCxcu9Ow/e3nn8ePH1bdvX5UoUcIT3tauXaujR4+qd+/envaGYWj//v1atWqVhg4d6jnfOnXqpB07dmT7XFJSUvTrr7/q1KlTevPNNz2vbd++XY0bN5bValWXLl3UuHFjtW7dWrVq1VJiYmKOrx88eNDT/8W+l+bNm3s+ryJFiujUqVOEPgA+h9AHAFfJuff0netsAJPcD8Lo0KGDnn/+ec/20aNHFR4eLj8/vywPJDl8+HCWB4RYLBZ9+umn2rp1q9atW6dx48YpKipKgwcPztJ/5j5cLpccDkeWWs8yDEOmaWrWrFme+6AiIyM1duzYLPWXK1dOsbGx6tmzp2rXrq1atWpd1nG88sorcjqdatOmjZo2barDhw/nOsjVrl1bM2bMkNPpzDLbt2XLFs2cOVMTJ06UpCz9ZWRkZOkj8/fwwAMPeC5HrVy5ssqVK6cdO3YoLCwsy3d57Ngxz0NMzienzzzz2JnHzQ2n05ntwTSmacrhcMjPzy/LMZ478ynl7jzJzGq1ZhsvPj5elSpVyvJa4cKF9cYbb6hdu3aqW7euWrVqJZfLpYYNG2aZyTx8+LCKFy+erdZzLyM++7m4XC6ZpqnZs2crODhYknT8+HEFBgYqNDRU8+fP12+//aaffvpJzz77rPr06aPu3bvn+HqTJk08/V/se8l8WfHZ/xYAwNewZAMA5IHGjRtr8eLFOnr0qCRp1qxZ6tWrlySpYcOG+vrrrz1PSHz66af166+/et67fft2tWvXTpUrV9Zjjz2m3r17a+vWrVn6j4qK0pdffumZQZs5c6ZuvfVWz/1XOXnwwQc1f/58zZ8/P1vgO6tevXrq2LGjRo4cKZfLdVnHsWbNGvXv319t27aVJG3evFlOpzNXn1vdunVVqVIljR8/Xna7XZI7kI0ZM0Zly5aV5A4lcXFxkqQ///wzy6zSuc7O3L399tvq0qWLJKlixYpZAvzhw4fVrl07T5/nExUVpVmzZikjI0Mul0ufffaZGjVqlKvjykmlSpXk5+en77//XpL7wUDfffed7rzzTjVp0kTff/+9Tp8+LUmaO3dutvdf6DyxWq1ZfglwdjzDMLR27VpJ0u+//65evXrJ5XJl67tcuXJ6/PHHNXbsWKWkpKhhw4Zau3atdu3aJUlatWqV7rvvPqWlpalJkyb68ssv5XK5lJqaqkWLFuX4lFWbzaY6depo+vTpktwzxg8++KCWLVumFStWqHfv3qpbt66eeuopdezYUXFxced9PbOr/b0AwI2ImT4AyAONGzdW37599fDDD8swDNlsNk2ZMkWGYejJJ5/U2LFj1aFDBzmdTrVt21atWrXS8uXLJUk1atRQmzZtFB0drZCQEAUFBemll17K0v/999+vw4cPq0uXLnK5XIqIiNCkSZOuSu0DBgxQmzZt9MUXX6hr166XfBwJCQnq37+/QkJCZLPZdOutt2r//v25Hv+tt97S66+/rs6dO8tqtcrlcqljx47q06ePJOmJJ57QkCFDtGrVKlWqVOmil1V26dJFU6dO9dwTGRAQoKlTp2rs2LH64IMP5HA49Mwzz6h+/foX7OeJJ57Qq6++qo4dO8rhcKhWrVoaPnx4ro7phx9+UN26dT3bYWFhWr16taZOnaoxY8Zo8uTJcjqd6t+/v+feywceeEAxMTEKCgpS1apVPbNjZ13oPLn77rv12muvZZnxCggI0OTJkzVu3DjFxsbK399fkydPPu8vCvr06aN58+Zp2rRpGjhwoEaNGqUBAwbINE35+flp2rRpCg0N1WOPPaZRo0apffv2CgsLU5EiRc67tMWkSZM0evRotW/fXunp6WrXrp3uu+8+OZ1OrV69Wu3atVNISIjCw8M1evRolSpVKsfXr9b3AgC+wjC5jgEAgBvK1q1btXHjRvXs2VOSNH36dG3evPmSHhRzrSxevFg2m01NmjSRy+XSU089pUaNGqlbt255XRoA5BuEPgAAbjBJSUkaNmyYdu/eLcMwVKpUKY0ePVolSpTI69KyiY+P14gRI5SamqqMjAzdfvvtGjZsmOfBKgAA7yP0AQAAAIAP8+qDXDZv3qwePXpke3358uWKjo5WTEyMvvjiC2+WAAAAAAD5mtce5PL+++9rwYIF2W4sz8jI0Pjx4zV37lwFBwfrwQcfVLNmzVSsWDFvlQIAAAAA+ZbXQl/58uU1efLkbOsB7dq1S+XLl1d4eLgkqX79+lq/fr3atGlz3r5cLpeSk5Pl7++f42OeAQAAAMCXmaapjIwMhYaGZlvz9GK8Fvpat26tgwcPZns9KSkpywK3oaGhSkpKumBfycnJio+Pv+o1AgAAAMCNpFq1alnyVG5c83X6bDabkpOTPdvJyckXLdrzhK+CJSQrSwvCt23566SOJKWquC1Ihgydndue/uufkqSqxQpIkgzJM/N9dgLc096Q532GcbYP9/+aMnUqNV2FQwLP9PHv2Jn7kefPmfcb52yfaW2c0+6cd2ap78yftx05pX+S02RICg8OyPLmzO0MzzszH1P2OjP3b/z7xiyvnbZnyM9iUUiANUttWY8qp88u0+vnHVtZrkTIUmOm9577eUrSP8l2FQ4JkMXIfJRZj/fcY89S7zljnVtX1rbnHm3mPs/93M/98zmfU6bNtXvci7OXDg+R1ZLz55ljrTm0y7HGTA1yOs6dCYnuz+9KLwa5Bo82O98VKzmVn5dPWsv8mDdL5i/g0nu6jD3eZzlznllk5LqOzF/dLaXcVyyZpiTT/XeraUrbjyZKkppWKfHv/jP7zo5jZto++zmfaebZl7nd2f1nn72XZf+ZPg+dStGxJLt63lpJKelOHTmdqjsrFpPVsMiQPP9dWi2GDMOQ9cxfN0F+fiocEuBeX9Fiyfr3MYDrSnp6uuLj4y/r6cfXPEFVrlxZ+/bt08mTJxUSEqL169d7FtQ9n7P/QLp6dJaOJWTd93A/Gd0fdu8f3F/a8HP2DmrXk+W19yRJ5hczZb77Zs7jLPlRRkCAzN07ZfZ9MOc2o/9Pxp1N3OM90Fr655/sjTo+IMtT7staXbEjpe8WZW9TtpwsM75217RsicxxOS8Ua3y2QEbJ0jJPnZTZuUXObZ4bKqNdtHu8x7tLO3dkbxTVTJaRE93jfTBF5qyPs7cJDpFl0Wp3m03rZQ58POfx3vpQxi213eO1vkNyOLK36dlXRq/H3G1efFb6aU32jm6uKcvk6e7xvp4tc0rOC0cbC1bJCA2VeWCvzN7359zm5Vdl3NXcPV739tLfh7M3urejLAPcCxO7XhsjLZ6XvU3JUrJ8ttBd0+plMl95IefxPp4ro1wFmcnJMu9rknObJwfJ6NTVPd5TD0nbtmZvdEdjWca+4R5vxrsyP3lfkZIiM7fx85Plu5/0WrliMn/fLPPpXjmP93/vyKjjXoTa1e4uKTUle5sHe8t45El3m5HPSz+syN5R1eqyvPOZu6ZFX8p8fXzO4321VEZ4QZl//yWz+305txk2WkZz96Xbrl6dpIMHPPs6nP1D63ayDB7pbjM5VpqXw8OdihSR5Yvv3DX9uErm8IE5j/f+LBmVqspMT5fZ5s6c2zz2jIwH3A+Ycg14VNr8W/ZG9W+XJfZt93iffSTzo6k59mVZtt7dJn6bzCd65jzehMkybm3oHq/T3VJiYvY293eT8cQAd5uxw6Tl32fvqGJlWT6Y4x7vu4UyY1/Jebw538goWlzmsaMyY9rm3GbwyzJat3eP90iMtGdX9kZ3t5LlxXHu8aa9JnPu59nbFCggy9fuBeM7O47KHPJUzuNN+0RGtZvd4zXPeaH0q/F3ucPlkmFI//x3hVx+/grYv0vhz/TOcbzkl16V8/YoSVJYr/ayHM/+d3lGu/uV3s99rgW+NkZ+Sxdna2OWKaf0D+dKkqwrvpN/7Igcx3PMmCejZGmdTjimkAfvORPwzxy73D/QL2ndU37t71eBIH9VG9FPBQ7uztbPyfqNtPcp9xil505XiUWzs7VxBQXr9/cXSJJs2zer4rhBWfafHXnPiLeUWu0WGYZUo1drGTn8Xf5PdC9tvydG87Ye0PPLP1CRLb/o3M5OV7pJW4e+JkNSqeXzVXHWO+46zKxhcf2UL2WGhCro74OqOeyRHD+nPf1f0slb3d/LLYN6KvDYkWxtjjdto4MPPyfDMFTmo9dVeOU32dpkFC2pP9/4VJKhsF9Wq9xb//734jKlJLt7UfpXe4xUQqGSqhUeoD7jcq5p3b29FHdHa8X9fVJPfhGrCn9n/17+qFhLH3foL5cptfppoVr+nP3ffafVT8OedP+9EvHXLvX7b2yO470TPUC7y1aXJI2e9owC09OytVne4B5926iTJKnH4ndU88+NWfYXkmQUL68B3YfLMKRbt6xSx2Wf5jjeuCdeV2qwTeGnjun5D4fl2MYybLT8W90rSQp+pIuMvw5k2W9Icra4V67nX5YkWd+eKMuC/3r2WSyGrIYh/2LFdcvqnyRJp1Ys1d7+j+U4XrWvFyu4eg250tO1tc5NObYp/fxQFXvI/Z3t6tVNSb9m/zsj7M7GqvTBDEnSkXen6u83/y/Hvmpvc/8dmBK3VTsf6Jhjm0rvTVdY47skSXEN68t56mS2NsV6PazSL7woSdo36Fmd/GZhtjZBVaup+vwlkqTjX3+pAy8OztZGkm5e8aP8S5RQxpEj2tYs53/Pyo2NVeFO7p//dnRoo7Sd2a+KK9i2vSImvSFJ+uvVsUqY8VG2Ntbwgopct0GSdHrNau1+9KEcx6v6xTyFRNaUJG2+uXKObUo+M1AlHusnSdr9SC+d/jH7z3+2W29X5Rnuf1MSpn+gvybm/HNGzU1/yBIQoNQd2xXf6d4c21R4+12FN3P/fLytyR3KSEjI1qZotx4q89JISdL+Yc/rxLyvsrUJjKigGkuWSZJOLFqg/YOfy3G8m/63SgFlyspx4oR+b5Tzv2dlR45RkQfcOSL+/vuUuu33bG3CW7RWhbfcP1scfn2ijr7/TrY2lpBQ1Vy/RZKU9MtP2tW7u8yixaTX376s292uWehbuHChUlJSFBMToyFDhqhPnz4yTVPR0dHX5bpCAIAbi9+Z+xtKhAW7f4EXGiTzPP8whgf5ywgNlCS5LEaOU6iB/lYFB7lnPVx+lpynWS2GQgPc/5SafhaZ55kNC/CzyvCzqnBIoEyrNcc2HSLLybilrHu8sCDJL3u7kgWCVbpicfd4hUJl5tBG/n66rXxRd5vj4ecdr0aJAjLKFnaPZ7Vmndo7o3SBYJWpWkrNq5aSa8Pnkn/2HxuCbUFqWa2Ue7zfC8r0y/lHi6hKJdy/wAtIzbluSTeXLCjjzPG5gvxz/AxKhYeoTIT74W+uAsFSDscXGOinemWLuGvaHZbtMwg60++k+xr8+wu813Ke4WpzUxm1vdv9qzjXug+l04fO7Pn3u46qVFxNujaSJCUnb1LA1oAs+w1Jpp+fXuvQQIZhyLItUCGLA3Mcb1DTm2XWbiBDUsBnwTJSs38v0bXKK/oB9y+TLH98LR0OlMs0ZXe45Gcx5HCZCvCzqkvtCJkyVeJEUYXk8N2lZjiUluFU7cqFleZK1vlW8fr8t73anOL+JdfAY6dVJNmetYEh/fbn3/ryK/cvBdr/8ZfuSLGf8ykZOu06pd6vL1bpAsGqvHOz2ialyeFyjxnkZ/X8JzZ1zXad3J2qlORUdUhN9/SQubpf4w5o35KNcplSg79OqFBKurIyFX/gH3V9fbFCAvzUd/dhlU/LyHZshqQpP2zX70dOqtqJQ6qflu6u5cylGWev4kh3unL8bIAbxQ2xTp/dbldcXJxUpIxkZTFXAAAAb/s7MVWDF21Q+1vKqkBgQKaLSd3+vUT139czX6a6cNtBpWU4s/VbNDRITtOUy2Uq0Z7hudXAkJFtjIsyc/xjFheaFMl8MfnZGs5ean22P0um2wBCAvxkPTNzeSrt36BZr2wR1StbWHaHU0l2h+6uWlLFbEGqVOTS7rsCLuRsJoqMjFRgYM6/ODofQh8AAADyhMs0lZiWIcOQXC5TTtMdIp0uUw6XyzODbxjyXBZ99r7frK+578/03DdsGLIY0mm7Q6dS0xXkb5Ur0z2Rmf9sytSGA8d1Mi1dP+87pjY3lZbVsMhpurTw938fSljcFqwgf4tcLsklU3+dSpEhQxZLpvu3MwXHs/eN+vtZZDHcQTElwyFD0p0Vi8uaw335mbePp9i15a8TalW9tCJLFXT3YTF0KjVDIQFWVS4S5h7jzFgWw5DLNFU2PEQWiyE/i/s1P4tFQf45z67jxkLoAwAAAK4x0zS18dBxBfpZZcj9IJ8NB//RgZP/PrSwStEwOU3J6XJp/4nkLPf05sQ4Z+Ns+7M/sZ8NeZ4f4M3MwVFZpikzP/AtLMA/UyjO+lAsh9Ol5HT3fb0BVqs61iwnP4t7jNqlC+nOCsUJjteBKwl9PAoTAAAAuAyGYXjuH5Xc96V2rlX+gu9JSXfIMJT1ia45zMGYpjRn015VLhomW6CfXC5TLlNa+effOmXP0B3li8qUe9bSNKXDian69cAxlQ0PVakCwXKdmTHd9NdxSVKRkEBPv+deRmt3uM7UZcjucGnm+t1njk+eGcbitiAlJLkfLFS7dGH1aFBJ5QuFqkJh2+V8dLjGmOkDAAAAoMS0DB1LTpPVYsjpMvXj3gR9v+MvSVKBIH+dTsvwPI1VZy5ZtRju8GpIqlw0TFbDUHxCoiIK21SzZEHFJyTK4TLVsnqpTPdMuv//n2S7qhQNU6OKxVU4JOCynkqZn3B5JwAAAACvcpmmvt/xl+bHHVSgn0XHU+yyBforyZ5x5v5GQy6Xy3Pfpams917mxGq41448+wCd8KAA3VW5hJpULqFgf6sMw1CpAsHu5T/OhMwgf6sKBOW/NSUJfQAAAACuC4lpGcpwuuQ0Tf2dmKojSWkqFRYsU6bn3kRTpuZs3KeDp9z3P559KI77ATj/3m9okc6ZATw7ryhPUGx3c1mlZDi1ZvdR/adBJYUFuu9g61SzvEICfOduNkIfAAAAgBuWw+XSoZMp+mHPUa3bm6CmVUrot4PHdehUiqIqlZBpuu9E3Hc8WQdPJcswDJmm6XkKqjJdNirpzJNL/w2LAVaret5aSX4Wi37/+6QeqBOh28+s+XmjIPQBAAAAyFcynC79k2yXxWIow+nSL/uPqVIRmw6dStXsjXuyLO+RORBKkr/FIn+rRQ6XSy2qldIDdSqoVqlCsliu3/sKCX0AAAAAcI7UDIeOp6Tr979Pqmx4iBwuU5NW/u5Zy9GQez3Dfy8tdbu5ZEHVL1tYt5QsqFIFQlQ4JEAlwoLz9FgIfQAAAACQS4lpGUpzODV38z79tC9Bks7MCv674P2/TxuVJyTWL1dEw1vWUunwkGteM6EPAAAAAK6AaZpKzXBq+9FErdlzVJv/Oq5CwQEqVSBEm/86LqthyGqxyM9q6OYSBfVhTEP5WS3XrD5CHwAAAAB4iWmamrtlnxb+flCGDAX6ue8JHND0Zt1fK+Ka3AtI6AMAAACAa2DZzsOauX635wmhVouhplVKKrZ9fa+OeyWh79rNRwIAAADADa551VJ6/M5qalqlhNIcTqVmOLVs52Hd9vpivbn6D12Pc2rM9AEAAADAZZq44ndtPXxCFsNQoJ9VFsO9MPzQFjWv6jjM9AEAAABAHni+2S36pFtj3V21pFIzHErNcOrLLft12+uLNW3tjuti5o+ZPgAAAAC4CkzT1KCFG3Qs2S6L4V4E3nJmGYjbI4rprU63yjAu76EvzPQBAAAAQB4zDEP/d18DvdvlDt1arqgKhQQoJcOhlAyn1u45qtvf+EY/7D5yzevyu+YjAgAAAIAPC/Kzql+j6p7tn/YlaNraHXJaLHpu3npZDGlQs1vUMKKYyhUK9Xo9hD4AAAAA8KI7IoqpVqlCil3xu/YeT5LVYmj80jhZLYZuKVlQM7o18ur4XN4JAAAAAF4WEuCnka1ra0jzSDWtXFLpTqdSMxzaduSk/jhyyqtjE/oAAAAA4BqpUTxcPRpU0ifdGqt8oVA5XKZ6fLZGA+b/6rUxCX0AAAAAkAceurWKyoSHKM3h1A+7j2ra2h1eGYfQBwAAAAB5oHLRML3SurYKBPrLnuHURz//qa+27L/q4xD6AAAAACAPvdXpVpmS7E6nJizbqm+3H7qq/RP6AAAAACAPGYahj7reKafLVLrTpZe+2aTH/rvuqvVP6AMAAACAPGYxDE2Nvl1VioYpzeHUnn+Srl7fV60nAAAAAMBlswX6a2jzmjIknUhJ1/oD/1yVfgl9AAAAAHAdCQ8KULrTpX5zf7oq/RH6AAAAAOA6Mum++nKapkxTijt84or7I/QBAAAAwHXE32pRgUB/pTtdenj2jzp6OvWK+iP0AQAAAMB15rUODeRwueRwmWr3wXIduYLgR+gDAAAAgOuMv9WiIc0jZXc45XCZOpWaftl9EfoAAAAA4Dp0c4mCal2jtExJr6/+47L7IfQBAAAAwHWqftkiynC6L/O8XIQ+AAAAALhO1SgertAAPx09nXbZfRD6AAAAAOA6lu5gpg8AAAAAfNaoe2rL4XJd9vsJfQAAAABwHSsdHiLDuPz3E/oAAAAAwIcR+gAAAADAhxH6AAAAAOB6d/nPcSH0AQAAAMD1rlHF4pf9XkIfAAAAAFznOtUsd9nvJfQBAAAAgA8j9AEAAACAD/Na6HO5XBoxYoRiYmLUo0cP7du3L8v+BQsWqFOnToqOjtbnn3/urTIAAAAAIF/z81bHS5cuVXp6uubMmaNNmzZpwoQJmjZtmmd/bGysFi1apJCQEN1777269957FR4e7q1yAAAAACBf8lro27Bhg6KioiRJderUUVxcXJb91atX1+nTp+Xn5yfTNGVcyRLzAAAAAIAceS30JSUlyWazebatVqscDof8/NxDVq1aVdHR0QoODlbLli1VoEABb5UCAAAAAPmW1+7ps9lsSk5O9my7XC5P4Nu+fbtWrlypZcuWafny5Tp+/LiWLFnirVIAAAAAIN/yWuirV6+eVq9eLUnatGmTqlWr5tkXFhamoKAgBQYGymq1qnDhwkpMTPRWKQAAAACQb3nt8s6WLVtq7dq16tq1q0zT1Lhx47Rw4UKlpKQoJiZGMTEx6tatm/z9/VW+fHl16tTJW6UAAAAAQL5lmKZp5nURF2O3290PgilSRrL653U5AAAAAHBtOTOkfw4pMjJSgYGBl/RWFmcHAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB9G6AMAAAAAH0boAwAAAAAfRugDAAAAAB/m562OXS6XRo4cqR07diggIEBjxoxRRESEZ/+WLVs0YcIEmaapYsWKaeLEiQoMDPRWOQAAAACQL3ltpm/p0qVKT0/XnDlzNHDgQE2YMMGzzzRNDR8+XOPHj9esWbMUFRWlQ4cOeasUAAAAAMi3vDbTt2HDBkVFRUmS6tSpo7i4OM++PXv2qGDBgpoxY4bi4+PVpEkTVapUyVulAAAAAEC+5bWZvqSkJNlsNs+21WqVw+GQJJ04cUIbN25Ut27dNH36dP30009at26dt0oBAAAAgHzLa6HPZrMpOTnZs+1yueTn555YLFiwoCIiIlSlShX5+/srKioqy0wgAAAAAODq8Froq1evnlavXi1J2rRpk6pVq+bZV65cOSUnJ2vfvn2SpPXr16tq1areKgUAAAAA8i2v3dPXsmVLrV27Vl27dpVpmho3bpwWLlyolJQUxcTEaOzYsRo4cKBM01TdunXVtGlTb5UCAAAAAPmWYZqmmddFXIzdbndf/lmkjGT1z+tyAAAAAODacmZI/xxSZGTkJS91x+LsAAAAAODDCH0AAAAA4MMIfQAAAADgw3L1IJdDhw7p008/1alTp5T5FsDx48d7rTAAAAAAwJXLVeh79tln1aBBAzVo0ECGYXi7JgAAAADAVZKr0OdwOPTCCy94uxYAAAAAwFWWq3v66tevr+XLlys9Pd3b9QAAAAAArqJczfR9++23+vTTT7O8ZhiG/vjjD68UBQAAAAC4OnIV+tasWePtOgAAAAAAXpCr0JeamqopU6Zo3bp1cjqduuOOO/TMM88oJCTE2/UBAAAAAK5Aru7pGzVqlFJTUzVu3Di9+uqrysjI0Msvv+zt2gAAAAAAVyhXM32///67FixY4NkeMWKE2rZt67WiAAAAAABXR65m+kzTVGJiomc7MTFRVqvVa0UBAAAAAK6OXM309e7dW/fff7/uvvtumaapFStW6NFHH/V2bQAAAACAK2SYpmnmpmF8fLx+/fVXuVwu3Xbbbapevbq3a/Ow2+2Ki4uTipSRrP7XbFwAAAAAuC44M6R/DikyMlKBgYGX9NYLXt65YsUKSdK8efO0bds2hYaGKiwsTH/88YfmzZt32fUCAAAAAK6NC17euXXrVjVr1kw///xzjvs7duzojZoAAAAAAFdJri/vPOv06dP6+++/VbVqVW/VlA2XdwIAAADI17x1eedZ//3vfzVkyBAdP35c9957r55++mm98847l1UrAAAAAODayVXomzVrlgYMGKBFixapefPmWrhwob7//ntv1wYAAAAAuEK5Cn2SVLx4ca1atUpNmzaVn5+f7Ha7N+sCAAAAAFwFuQp9VapU0WOPPaaDBw+qYcOGevbZZ1WzZk1v1wYAAAAAuEK5Wpx93Lhx2rhxo6pWraqAgADdd999atKkibdrAwAAAABcoQuGvjlz5igmJsbz0JbMSzds27ZNTz75pHerAwAAAABckQte3nmJqzkAAAAAAK4zuVqnz+FwaNWqVWrevLmOHz+u5cuXKzo6WoZhXIsaWacPAAAAQP7m7XX6hg8fnmWJhp9//lkvv/zypRUJAAAAALjmcvUgl7i4OC1cuFCSVLhwYU2cOFHt27f3amEAAAAAgCuXq5k+l8ulo0ePerb/+ecfWSy5XuIPAAAAAJBHcjXT9/jjj6tTp06qX7++JGnz5s168cUXvVoYAAAAAODK5epBLpJ05MgRbdq0SX5+fqpZs6aKFy/u7do8eJALAAAAgHzN2w9ySU9P19dff61ly5bptttu0xdffKH09PTLqhUAAAAAcO3kKvSNGjVKKSkp2rZtm/z8/LR//34NGzbM27UBAAAAAK5QrkLf77//rgEDBsjPz0/BwcF69dVXtX37dm/XBgAAAAC4QrkKfYZhKD093bMY+4kTJ67ZwuwAAAAAgMuXq6d39uzZUw899JASEhI0duxYLV26VP379/d2bQAAAACAK5Sr0HfXXXcpMjJSP//8s5xOp6ZNm6YaNWp4uzYAAAAAwBXKVejr3r27lixZoipVqni7HgAAAADAVZSr0FejRg3NmzdPtWrVUlBQkOf10qVLe60wAAAAAMCVy1Xo27x5s7Zs2aLM67gbhqFly5Z5rTAAAAAAwJW7YOg7cuSIYmNjFRoaqrp162rQoEEqUKDAtaoNAAAAAHCFLrhkw7Bhw1S8eHENHDhQGRkZGj9+/LWqCwAAAABwFVx0pu/DDz+UJDVq1EgdO3a8FjUBAAAAAK6SC870+fv7Z/lz5m0AAAAAwPXvgqHvXIZheKsOAAAAAIAXXPDyzp07d6p58+ae7SNHjqh58+YyTZOndwIAAADADeCCoe+77767VnUAAAAAALzggqGvTJkyl92xy+XSyJEjtWPHDgUEBGjMmDGKiIjI1m748OEKDw/XoEGDLnssAAAAAEDOLumevkuxdOlSpaena86cORo4cKAmTJiQrc3s2bMVHx/vrRIAAAAAIN/zWujbsGGDoqKiJEl16tRRXFxclv0bN27U5s2bFRMT460SAAAAACDf81roS0pKks1m82xbrVY5HA5J0tGjRzVlyhSNGDHCW8MDAAAAAHSRe/quhM1mU3Jysmfb5XLJz8893LfffqsTJ07o0UcfVUJCgtLS0lSpUiV17tzZW+UAAAAAQL7ktdBXr149rVixQm3bttWmTZtUrVo1z76ePXuqZ8+ekqSvvvpKu3fvJvABAAAAgBd4LfS1bNlSa9euVdeuXWWapsaNG6eFCxcqJSWF+/gAAAAA4BoxTNM087qIi7Hb7e4HwRQpI1n987ocAAAAALi2nBnSP4cUGRmpwMDAS3qr1x7kAgAAAADIe4Q+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYX7e6tjlcmnkyJHasWOHAgICNGbMGEVERHj2L1q0SDNmzJDValW1atU0cuRIWSxkUAAAAAC4mryWspYuXar09HTNmTNHAwcO1IQJEzz70tLS9MYbb+iTTz7R7NmzlZSUpBUrVnirFAAAAADIt7wW+jZs2KCoqChJUp06dRQXF+fZFxAQoNmzZys4OFiS5HA4FBgY6K1SAAAAACDf8lroS0pKks1m82xbrVY5HA73oBaLihYtKkmaOXOmUlJS1KhRI2+VAgAAAAD5ltfu6bPZbEpOTvZsu1wu+fn5ZdmeOHGi9uzZo8mTJ8swDG+VAgAAAAD5ltdm+urVq6fVq1dLkjZt2qRq1apl2T9ixAjZ7XZNnTrVc5knAAAAAODqMkzTNL3R8dmnd8bHx8s0TY0bN07btm1TSkqKIiMjFR0drQYNGnhm+Hr27KmWLVvm2JfdbnffE1ikjGT190a5AAAAAHD9cmZI/xxSZGTkJT8PxWuh72oi9AEAAADI164g9LEwHgAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4ML+8LuCKuZySaeZ1FfA2w5As1ryuAgAAALjh3NChL8iVrmIFQhXgf0MfBnIhPcOhhMRkpVkC8roUAAAA4IZy46Yll1PFCoQqNDQkryvBNeAf4A57BxLTmPEDAAAALsGNe0+faTLDl88E+PtxKS8AAABwiW7c0AcAAAAAuCimyq7A+l9/0QuDBqpSpcoyTVMOh0MvjxqtipUqaeKr4/Wfnr1UqlRpr41/W93aql2nrgxDcjgcqlipkoa+NELfLFqod96eojJly0mSTp9OVO06dTX0peFeqwUAAADA9YnQd4Vuve12TZg4SZK07se1ev3/Jumtt6fq+ReGen3s8PBwvT/9Y8/2C4MGau0PqyVJ97S9V08/N0CS5HK51KdXD237PU433xLp9boAAAAAXD98JvRNXbtDK//8+6r22bRKSfVrVD3X7RMTE1W6tHtmr+9DvTVs+Ah99+0S/XXooI7/c1yHD/+lgYNf0J2NGmvp99/pi9mzPLeoxb72unb9uVNvvf6a/Pz9dfsdDbVm9SrNnDVHkjvQ9ezdW7dE1sxx7IyMDKWmpCg4JESnTp3Ksi85OVmnT5+WzRaW5fX9+/Zp1MsjlJGRoaDgIE2InaQ3/m+SWrVpo0aNo7R2zQ/6fskSvTJ2nNq2aqEKFSuqXLnyWvfjWs2Z+5WCQ0I0Y/pHslqtatGqlcaMHKn0dLsCAgL10siRKlmyVK4/OwAAAADe4TOhL6/8+svP6vtQb6Wnp2vnzni99fbUbG38/QM05Z139dOPP2rmJx/rzkaNtW/fPr359jQFBwdrzCsjtW7tWhUvUVx2u12ffD7b3ffPP2v3rj9VpGhR/XXoYLbAd+rUKfV9qLcMQ5JhqFHjKN12+x1aMO9rffvNYm3ZvFnHjiUoJCRUffo+qvIREVne//qkiXrokUfUqHGUvv/2W23f/sd5j/PI33/r8y/mqmDBgnrztf/TsqX/U7v7Oui7JUs09b33NX7MaD3Y/T9qFBWln3/6SZNff11jX4298g8YAAAAwBXxmdDXr1H1S5qVu1oyX965d88e9f5PN327bEWWNjVuukmSVKJkSaXb0yVJhQsX1ssvDlNwSIj27tmjWrVrS5IqVKjoeV+n6Pu1YP48lSxZSm3btc829rmXd2Z29vLOQwcP6sknHlNEhQrZ2uzbu1e1ateRJLW65x5J0reLF//bINODMgsWKqSCBQt66ho3epQqVKyoiAoRKliwoP7cGa8PP3hPH3/0oUzTlL+/f84fGAAAAIBrymdC3/WgSJEiOb5uGEaW7dOnT+udqW/rm++XSpKeePQRz2WehuXfB6q2aNVKM2dMV3jBgor9v9cuq6YyZctqyLCXNHjgAP336/kKDg727KtYqZK2xcXp9oYN9c2iRUo8dUoBgYE6lnBMkvTHH9s8bS2ZjqF8RIRMmfpk+nTdHxMjSapQsZJ69u6t2nXqas/u3dqwfv1l1QsAAADg6iL0XaGzl3darBalJCdrwPODFRQUdMH32Gw21alTV90euF/BwSEqUKCAEhKOqkzZMlnaBQYGql79Bjpx4rjCwwtedo23N2yo2++4Q+9MfVvPDRzkef2ZgQM1dtQr+uC9dxUUFKQxEybo0MGDGjl8uJYsXqSIChHn7bNjp86aNmWKbr3tdknScwMHadyYUUq3p8tuT9Oga/AgGwAAAAAXZ5jm9b/atd1uV1xcnFSkjGQ9c9mg06HKRcPkHxCQt8V52fgxo9S8ZSvddvsdeV1KnstIT9euY6clK7+rAAAAQD7jzJD+OaTIyEgFBgZe0ltZnP061u/RvkpLTSPwAQAAALhsTJlcx6a+935elwAAAADgBsdMHwAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4MELfVfLxhx+oVbMmstvteV1KNmvX/KAv//vFefcPfPaZqzbWiRMn1O/Rvnq4Vw+9MGigUlNTs7WZOGG8usc8oL4P9dbWLVskSYcOHlSfXj31cK8eenHICzm+DwAAAMClI/RdJd8sXqxW97TRd0u+yetSsmnUOErRXR447/7/e+PNqzbW++9M0z1t2+qjGTNVo0aNbGFz9aqV2rt3j2bOmq3Y117XhLFjJElvvDZJ9z/wgD6aMVMNbr1Vn30y46rVBAAAAORnPrVkw87a1XJ8vchTA1T4kcclSYcef0gp69ZmaxPc4HaV/XCmJOnEjA917LVXVXVzfK7GXf/rLypXrpzufyBGLw19QTVuulmTYifovQ+nS5Ke7t9P/Z58SkePHtE7b0+RzWZTWIFwVa1WTY/3659jn30f6q0KFStq7549Mk1TEyZN0t49e/TW66/Jz99fne/volIlS+ntyW/KYrGqbLlyenHEy3I6nRo5/CUd/usvORwODR46TPv27tHePXv0WL/+emHQACWdTpLdnqZnBgxUg1tvU8umd+l/K1dr+x9/KHb8WFksVgUEBmj4yFdkukwNHfy8SpYsqQMHDiiyZk0NGz5Csz//TMv+978sNY8aN06bNv6mh/v2lSQ1iorSlDff1H969vK02b1rlxre2UgWi0WFChWS1WrRsWMJ2r1rl4a//IokqXbduvq/2Fdz9dkDAAAAuDCfCn155esvv1TH6GhVqFhRAQEBstvtsqfZ9ddff8nf318nT55Q1WrV9Pxzz+rjTz9TkaJF9eILgy/ab+06dfTiiJf1xexZ+uj993V3ixay2+365PPZMk1Tndrfq49mzFThIkU0dfJbWjh/nlJSUlS6dGlNmDhJf+7cqZ9/WqewsDBJ0sEDB/TPsWOa9v6HOn78H+3fuy/LeKNHvqwRr7yi6jVu0srly/XaxFg9N/B57d+3V1Pfe19BQUFq36a1jh1LUNdu3dW1W/dsNSclJclmc48XEhKqpKTTWfZXr1FDM2d8rJgHu+nI339r159/KjU1VdWr19CqlSvUvkNHrV6xgss7AQAAgKvEp0Jfbmbmyrwz/aJtCvXqo0K9+uRqzMRTp7T2h9U6cfy4Zn/+mZJOJ2nOrM/VsXNnLV4wX/4BAbqvQyedOHFcobZQFSlaVJJUt359HTt27IJ933rb7ZLc4W/liuWSpAoVKkqSThw/rmMJCXph0EBJkt2epjsa3qkTJ06oUeMoSVKVqlVVpWpVLZj3tSSpcpUqeqDrgxo2+Hk5HA517Z41tCUkHFX1GjdJkurVr6+33nhdklSufHmFhoZKkooWK6Z0e/p5Z/psNptSkpMVFBSklJRk2cIKZGnT8M5G+j0uTo898rCqVauum26+RQXDC+q55wfr1XFj9O2Sb3Tb7XeoUMFCufr8AQAAAFyYT4W+vPDNokXq0Dlazw0cJElKTU1V+3ta68lnntHgAc9JhqGp776vkJAQpSSn6MTx4ypUuLC2bt6sUmXKXLDvP7ZtU4mSJbVp40ZVrlxFkmRY3LdhFixUSMVLlNRrb01WWFiYVq1YruCQEP25c6d+j9uqpnffrYMHDmjq5Ld0x513SpJ2xscrOTlZb02dpoSEBD3Uo7vuatLUM16xYsUVv2OHqlWvrg3r1ysiIsK9wzCy1Xa+mb7adepqzQ+rdV/HTlr7ww+qV69elv379u5V4cKF9dGMmfr778MaMWyowgoU0KoF8/Xo4/1UrXp1zZzxsW5v2DB3XwAAAACACyL0XaGvv5qr0eMmeLaDg4PVvGULLV64UFWrVZfT6ZTNZpMkvTDsRT3V73HZbGFymS6VOxOq+j3aV2++/bb8/QOy9L1w/jx9+skMBQcHa/T4Cfpz578zmRaLRc8PGaJn+j8hl8tUqC1Uo8eOV+06dfXK8Jf0SO9ecrmcGvTCEP25c6ckqXxEhN57Z6oWL1wgf39/Pd7/ySzjDR/5il4dN1amTPlZrRrxyuhL/jweefQxjXhpmL7+cq4KFiykca/GSnI/qKVFy1aqWq26fly7RvO+/kqBAYEa8uJLkqSIChX1yojhCgjwV6XKVTyvAwAAALgyhmmaZl4XcTF2u11xcXFSkTKS1d/9otOhykXD5B8QcOE3X0c++uB9/adnLwUEBOjFIS+o4Z13qt19HXJs2/eh3ho2fIQqVqp0jau8fmWkp2vXsdOSld9VAAAAIJ9xZkj/HFJkZKQCAwMv6a389HwNhYSEqFf3BxUUFKTSpcuo1T1t8rokAAAAAD6OmT7cMJjpAwAAQL51BTN9LM4OAAAAAD7sxg19hqH0DEdeV4FrKD3DkeOTRAEAAACc3417nZzFqoTEZElSgP+NexjInfQMh/v7tnA5LwAAAHApbui0lGYJ0IHENOn6vy0RV8owCHwAAADAZbihQ58kyWLN6woAAAAA4LrltXv6XC6XRowYoZiYGPXo0UP79u3Lsn/58uWKjo5WTEyMvvjiC2+VAQAAAAD5mtdC39KlS5Wenq45c+Zo4MCBmjBhgmdfRkaGxo8fr48++kgzZ87UnDlzlJCQ4K1SAAAAACDf8trlnRs2bFBUVJQkqU6dOu519s7YtWuXypcvr/DwcElS/fr1tX79erVpk/Ni5WeXEgwwJIuF+/cAAAAA5C8ul5Suf7PRpfBa6EtKSpLNZvNsW61WORwO+fn5KSkpSWFhYZ59oaGhSkpKOm9fGRkZkqT0Y4e8VS4AAAAAXPcyMjIUFBR0Se/xWuiz2WxKTk72bLtcLvn5+eW4Lzk5OUsIPFdoaKiqVasmf39/GazTBgAAACCfMU1TGRkZCg0NveT3ei301atXTytWrFDbtm21adMmVatWzbOvcuXK2rdvn06ePKmQkBCtX79effr0OW9fFovlgqEQAAAAAHzdpc7wnWWYl3NRaC64XC6NHDlS8fHxMk1T48aN07Zt25SSkqKYmBgtX75cb7/9tkzTVHR0tLp37+6NMgAAAAAgX/Na6AMAAAAA5D2vLdkAAAAAAMh7hD4AAAAA8GHXXehzuVwaMWKEYmJi1KNHD+3bty/L/uXLlys6OloxMTH64osv8qhK5EcXOzcXLVqkLl26qGvXrhoxYoRcLlceVYr85mLn5lnDhw/XpEmTrnF1yO8udn5u2bJF3bp104MPPqinn35adrs9jypFfnOxc3PBggXq1KmToqOj9fnnn+dRlcjPNm/erB49emR7/XLy0HUX+pYuXar09HTNmTNHAwcO1IQJEzz7MjIyNH78eH300UeaOXOm5syZo4SEhDysFvnJhc7NtLQ0vfHGG/rkk080e/ZsJSUlacWKFXlYLfKTC52bZ82ePVvx8fF5UB3yuwudn6Zpavjw4Ro/frxmzZqlqKgoHTrEmry4Ni72d2dsbKymT5+uWbNmafr06Tp16lQeVYr86P3339dLL72U7Rdhl5uHrrvQt2HDBkVFRUmS6tSpo7i4OM++Xbt2qXz58goPD1dAQIDq16+v9evX51WpyGcudG4GBARo9uzZCg4OliQ5HA4FBgbmSZ3Ify50bkrSxo0btXnzZsXExORFecjnLnR+7tmzRwULFtSMGTP0n//8RydPnlSlSpXyqlTkMxf7u7N69eo6ffq00tPTZZoma0XjmipfvrwmT56c7fXLzUPXXehLSkqSzWbzbFutVjkcDs++zOv1hYaGKikp6ZrXiPzpQuemxWJR0aJFJUkzZ85USkqKGjVqlCd1Iv+50Ll59OhRTZkyRSNGjMir8pDPXej8PHHihDZu3Khu3bpp+vTp+umnn7Ru3bq8KhX5zIXOTUmqWrWqoqOjde+996pp06YqUKBAXpSJfKp169by88u+pPrl5qHrLvTZbDYlJyd7tl0ul+eAz92XnJzMou24Zi50bp7dfvXVV7V27VpNnjyZ3wjimrnQufntt9/qxIkTevTRR/Xee+9p0aJF+uqrr/KqVORDFzo/CxYsqIiICFWpUkX+/v6KiorKNtsCeMuFzs3t27dr5cqVWrZsmZYvX67jx49ryZIleVUq4HG5eei6C3316tXT6tWrJUmbNm1StWrVPPsqV66sffv26eTJk0pPT9f69etVt27dvCoV+cyFzk1JGjFihOx2u6ZOneq5zBO4Fi50bvbs2VNfffWVZs6cqUcffVTt2rVT586d86pU5EMXOj/LlSun5ORkzwM01q9fr6pVq+ZJnch/LnRuhoWFKSgoSIGBgbJarSpcuLASExPzqlTA43LzUPY5wzzWsmVLrV27Vl27dpVpmho3bpwWLlyolJQUxcTEaMiQIerTp49M01R0dLRKlCiR1yUjn7jQuRkZGam5c+eqQYMG6tWrlyT3D9stW7bM46qRH1zs700gL13s/Bw7dqwGDhwo0zRVt25dNW3aNK9LRj5xsXMzJiZG3bp1k7+/v8qXL69OnTrldcnIx640DxmmaZrXoE4AAAAAQB647i7vBAAAAABcPYQ+AAAAAPBhhD4AAAAA8GGEPgAAAADwYYQ+AAAAAPBh192SDQAAeNPBgwd1zz33qHLlypLcCzInJyerY8eOevrpp6/KGJMnT5YkPfXUU6pevbp27NhxVfoFAOByEPoAAPlO8eLFNX/+fM/2kSNH1Lp1a917772eMAgAgK/g8k4AQL6XkJAg0zQVGhqq9957T506ddJ9992n2NhYnV3O9uOPP1br1q3Vtm1bTZw4UZIUHx+vHj16KDo6Ws2aNdOsWbPy8jAAAMgRM30AgHzn6NGj6tChg+x2u06cOKGaNWtqypQpio+PV1xcnObOnSvDMPT8889rwYIFqlixoj7//HN9+eWXCg4O1iOPPKK4uDjNnz9f/fr1U8OGDXXgwAHdd999evDBB/P68AAAyILQBwDId85e3ulyuTRhwgTt2rVLjRo10sSJE7VlyxZ17txZkpSWlqbSpUvr2LFjatasmcLCwiS5Z/0k6aabbtIPP/ygd999V/Hx8UpJScmrQwIA4LwIfQCAfMtisWjw4MHq2LGjPvzwQzmdTvXq1UsPPfSQJCkxMVFWq9Uz83fWkSNHFBwcrBdffFEFChRQs2bN1LZtWy1atCivDgUAgPPinj4AQL7m5+enwYMHa+rUqbr55ps1f/58JScny+FwqH///vruu+/UoEEDrVq1yvP6wIEDFRcXp7Vr1+rpp59WixYttHr1akmS0+nM4yMCACArZvoAAPneXXfdpbp162r9+vVq1aqVHnjgATmdTkVFRalTp04yDEP/+c9/1LVrV7lcLrVs2VJ33nmnnnrqKXXr1k2BgYGqUaOGypQpo4MHD+b14QAAkIVhnn0sGQAAAADA53B5JwAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4MEIfAAAAAPgwQh8AAAAA+DBCHwAAAAD4MEIfAAAAAPiw/wepEJM7cuz2BQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Precision-Recall Curve for LogisticRegression'}, xlabel='Recall', ylabel='Precision'>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from yellowbrick.classifier import PrecisionRecallCurve, DiscriminationThreshold\n",
    "fig, ax = plt.subplots(figsize=(15,5))\n",
    "viz = PrecisionRecallCurve(lr)\n",
    "viz.fit(X_train, y_train)\n",
    "viz.score(X_test, y_test)\n",
    "viz.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4016ed4",
   "metadata": {},
   "source": [
    "# Thresholding <a class=\"anchor\" id=\"thresholding\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48e3c12d",
   "metadata": {},
   "source": [
    "## Classification Report of Default Threshold<a class=\"anchor\" id=\"cr\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "0122eb0e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.92      0.61      0.73       691\n",
      "           1       0.91      0.99      0.95      2698\n",
      "\n",
      "    accuracy                           0.91      3389\n",
      "   macro avg       0.91      0.80      0.84      3389\n",
      "weighted avg       0.91      0.91      0.90      3389\n",
      "\n"
     ]
    }
   ],
   "source": [
    "### Classification Report of current threshold ###\n",
    "\n",
    "y_pred = lr.predict(X_test)\n",
    "class_report = classification_report(y_test,y_pred)\n",
    "\n",
    "print(class_report)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "daabc7da",
   "metadata": {},
   "source": [
    "## Visualize accuracy with Confusion Matrix <a class=\"anchor\" id=\"cm\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "50496c74",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The recall score is: 0.99\n",
      "The precision score is: 0.91 \n",
      "\n",
      "Accuracy score is: 0.91\n",
      "The F1 score is: 0.95\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Get Confusion Matrix ###\n",
    "y_pred = lr.predict(X_test)\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "\n",
    "### Visualize Confusion Matrix ###\n",
    "groupNames = [\"True Neg\", \"False Pos\", \"False Neg\", \"True Pos\"]\n",
    "groupCount = [\"{0:0.0f}\".format(value) for value in\n",
    "                cm.flatten()]\n",
    "groupPercent = [\"{0:.2%}\".format(value) for value in\n",
    "                     cm.flatten()/np.sum(cm)]\n",
    "labels = [f\"{v1}\\n{v2}\\n{v3}\" for v1, v2, v3 in\n",
    "          zip(groupNames,groupCount,groupPercent)]\n",
    "labels = np.asarray(labels).reshape(2,2)\n",
    "\n",
    "sns.heatmap(cm, annot=labels, fmt='', cmap='rocket', annot_kws={\"fontsize\":20})\n",
    "plt.title(\"Confusion Matrix\")\n",
    "\n",
    "print(\"The recall score is: %.2f\" % recall_score( y_test, y_pred))\n",
    "print(\"The precision score is: %.2f\" % precision_score( y_test, y_pred),\"\\n\")\n",
    "print(\"Accuracy score is: %.2f\" % accuracy_score( y_test, y_pred))\n",
    "print(\"The F1 score is: %.2f\" % f1_score( y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acabb815",
   "metadata": {},
   "source": [
    "# Best Threshold Options "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47166b48",
   "metadata": {},
   "source": [
    "## Optimize Balanced Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "68e86526",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Balance Accuracy: 0.79724\n",
      "Threshold=0.804, Balance Accuracy=0.89194\n"
     ]
    }
   ],
   "source": [
    "y_pred = lr.predict(X_test)\n",
    "score = balanced_accuracy_score(y_test, y_pred)\n",
    "print('Balance Accuracy: %.5f' % score)\n",
    "\n",
    "# predict probabilities\n",
    "y_pred = lr.predict_proba(X_test)\n",
    "\n",
    "# keep probabilities for the positive outcome only\n",
    "probs = y_pred[:, 1]\n",
    "\n",
    "# define thresholds > test between between 0.0 and 1.0\n",
    "thresholds = np.arange(0.0, 1.0, 0.001)\n",
    "\n",
    "# apply threshold to positive probabilities to create labels\n",
    "def to_labels(pos_probs, threshold):\n",
    "    return (pos_probs >= threshold).astype('int')\n",
    "\n",
    "# evaluate each threshold\n",
    "scores = [balanced_accuracy_score(y_test, to_labels(probs, t)) for t in thresholds]\n",
    "\n",
    "# get best threshold\n",
    "ix = np.argmax(scores)\n",
    "print('Threshold=%.3f, Balance Accuracy=%.5f' % (thresholds[ix], scores[ix]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9be479a2",
   "metadata": {},
   "source": [
    "## Optimize F1 Score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5ad4061d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F-Score: 0.94548\n",
      "Threshold=0.664, F-Score=0.95271\n"
     ]
    }
   ],
   "source": [
    "y_pred = lr.predict(X_test)\n",
    "score = f1_score(y_test, y_pred)\n",
    "print('F-Score: %.5f' % score)\n",
    "\n",
    "# predict probabilities\n",
    "y_pred = lr.predict_proba(X_test)\n",
    "\n",
    "# keep probabilities for the positive outcome only\n",
    "probs = y_pred[:, 1]\n",
    "\n",
    "# define thresholds > test between between 0.0 and 1.0\n",
    "thresholds = np.arange(0.0, 1.0, 0.001)\n",
    "\n",
    "# apply threshold to positive probabilities to create labels\n",
    "def to_labels(pos_probs, threshold):\n",
    "    return (pos_probs >= threshold).astype('int')\n",
    "\n",
    "# evaluate each threshold\n",
    "scores = [f1_score(y_test, to_labels(probs, t)) for t in thresholds]\n",
    "\n",
    "# get best threshold\n",
    "ix = np.argmax(scores)\n",
    "print('Threshold=%.3f, F-Score=%.5f' % (thresholds[ix], scores[ix]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "480cfe39",
   "metadata": {},
   "outputs": [],
   "source": [
    "def precision_recall(model, X_test, y_test, threshold, beta=1):\n",
    "    \n",
    "    y_pred = (model.predict_proba(X_test)[:,1] >= threshold).astype(bool)\n",
    "    cm = confusion_matrix(y_test, y_pred)\n",
    "    \n",
    "    fp = cm[0,1]\n",
    "    tp = cm[1,1]\n",
    "    fn = cm[1,0]\n",
    "    tn = cm[0,0]\n",
    "    \n",
    "    precision = tp/(tp+fp)\n",
    "    recall = tp/(tp+fn)\n",
    "    \n",
    "    bal_acc=balanced_accuracy_score(y_test,y_pred)\n",
    "    fbeta = fbeta_score(y_test, y_pred, beta=beta)\n",
    "    \n",
    "    return([precision, recall, bal_acc, fbeta])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "54542157",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/4s/dkl5yc9543jbyd5lrgf388vh0000gp/T/ipykernel_96525/2530030121.py:11: RuntimeWarning: invalid value encountered in long_scalars\n",
      "  precision = tp/(tp+fp)\n"
     ]
    }
   ],
   "source": [
    "threshold_df = pd.DataFrame(columns=['Precision','Recall', 'Balanced Accuracy', 'F1-Measure'])\n",
    "\n",
    "for threshold in np.arange(0.0, 1.0, 0.001):\n",
    "    \n",
    "    threshold_df.loc[threshold] = precision_recall(lr, X_test, y_test, threshold)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed36750c",
   "metadata": {},
   "source": [
    "## Business Case 1 - Track Customer Sentiment - Balanced Accuracy\n",
    "\n",
    "Airlines can use this model to track overall customer sentiment on Twitter. This can help them understand how their company is perceived overall, if there are any upticks in general negative sentiment against the airline that they may need to address, and could help them respond more quickly to such situations. In this scenario, there will be an emphasis on balanced accuracy because, since we are trying to get an overall accurate representation of sentiment, all classes have equal importance. The managers then can analyze the tweets and be able to make an assessment on what needs to be improved for the company so they improve customer relations and not lose money.\n",
    "\n",
    "The default threshold is 0.5, but changing the threshold can make a better balanced accuracy score. Having a threshold of 0.70 allows the model to have higher balanced accuracy. Balanced accuracy is a good metric to use for this scenario because the data is imbalanced, and we care about the positive and negative classes.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "cc9123c4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = threshold_df.plot(figsize=(15,9))\n",
    "ax.set_xlabel('Threshold')\n",
    "ax.set_title('Precision-Recall vs. Threshold: Balanced Accuracy')\n",
    "\n",
    "ax.axvline(0.70, color='black', alpha=0.3)\n",
    "ax.text(0.686,.7,'Optimal Threshold',rotation=90, alpha=0.8, fontsize=16)\n",
    "\n",
    "ax.axvline(0.5, color ='black', alpha = 0.3)\n",
    "ax.text(.485,.7,'Default Threshold',rotation=90, alpha=0.8, fontsize=16);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "62f5c187",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The recall score is: 0.95\n",
      "The precision score is: 0.95 \n",
      "\n",
      "Accuracy score is: 0.92\n",
      "The F1 score is: 0.95\n",
      "The balanced accuracy is: 0.88\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictions = lr.predict_proba(X_test)[:,1] >= 0.70\n",
    "\n",
    "print(\"The recall score is: %.2f\" % recall_score( y_test, predictions))\n",
    "print(\"The precision score is: %.2f\" % precision_score( y_test, predictions),\"\\n\")\n",
    "print(\"Accuracy score is: %.2f\" % accuracy_score( y_test, predictions))\n",
    "print(\"The F1 score is: %.2f\" % f1_score( y_test, predictions))\n",
    "print(\"The balanced accuracy is: %.2f\" % balanced_accuracy_score(y_test, predictions))\n",
    "\n",
    "groupNames = [\"True Neg\", \"False Pos\", \"False Neg\", \"True Pos\"]\n",
    "groupCount = [\"{0:0.0f}\".format(value) for value in\n",
    "                cm.flatten()]\n",
    "groupPercent = [\"{0:.2%}\".format(value) for value in\n",
    "                     cm.flatten()/np.sum(cm)]\n",
    "labels = [f\"{v1}\\n{v2}\\n{v3}\" for v1, v2, v3 in zip(groupNames,groupCount,groupPercent)]\n",
    "labels = np.asarray(labels).reshape(2,2)\n",
    "\n",
    "cm = confusion_matrix(y_test , predictions )\n",
    "sns.heatmap(cm, annot=labels, annot_kws={\"size\": 20}, fmt=\"\", cmap=\"rocket\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59d62214",
   "metadata": {},
   "source": [
    "## Business Case 2 - Customer Service\n",
    "\n",
    "Airlines can also use this model to identify customers that are having a negative experience and direct tweets with negative sentiment towards the proper channels. The customer then can be helped in a timely manner which may improve the overall customer experience with the airline and the perception of the airline. \n",
    "\n",
    "One way this might be accomplished would be to set up a bot to alert customers with a message if it is a positive sentiment like \"Thank you for flying with us. We look forward to seeing you on your next trip.\" If it is a negative sentiment, a bot can send a similar tweet saying \"Sorry for the unsatisfactory experience. Can you please dm us.\" The bot can then send an alert to a customer service representative, and a discount or refund can be offered to the customer.\n",
    "\n",
    "In this scenario, a false negative would mean that the model predicted a tweet to be positive, but it was actually negative. The down side to this is that the customer will not receive the proper customer service help and may never fly with a particular airline again. This will ultimately cause the airline company to lose money. A false positive in this scenario would mean that the model predicted a negative sentiment tweet, but it was actually a positive sentiment tweet. The down side to this is that a bot would respond to a customer that had a positive experience and offer a discount or refund and cause a customer service representative to reach out when there isn’t a reason to. This will cause the company to lose money and waste time.\n",
    "\n",
    "In this scenario, since both low recall and low precision have significant downsides, we could use the F1-score. We want to find as much as possible of the negative sentiment tweets. We also don't need bots and customer service representatives wasting time and money on responding to customers and offering discounts if the tweet is a positive sentiment. This metric might change based on an individual airline’s particular situation in which either finding more of the customers with negative sentiment or saving more money would take on more importance. \n",
    "The default threshold and the optimized F1-score have similar metrics. We would need a cost analysis of the false positives and compare them to the false negatives to determine the next best step for the airline. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "60687744",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = threshold_df.plot(figsize=(15,9))\n",
    "ax.set_xlabel('Threshold')\n",
    "ax.set_title('Precision-Recall vs. Threshold: F1-Mesure')\n",
    "ax.axvline(0.664, color='black', alpha=0.3)\n",
    "ax.text(0.650,.7,'Optimal Threshold',rotation=90, alpha=0.8, fontsize=16)\n",
    "ax.axvline(0.5, color ='black', alpha = 0.3)\n",
    "ax.text(.485,.7,'Default Threshold',rotation=90, alpha=0.8, fontsize=16);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "7c52be39",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The recall score is: 0.96\n",
      "The precision score is: 0.95 \n",
      "\n",
      "Accuracy score is: 0.92\n",
      "The F1 score is: 0.95\n",
      "The balanced accuracy is: 0.87\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcYAAAFJCAYAAAABhI0kAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAABY3UlEQVR4nO3ddVhVyxrA4R+tEoLdDTZ2KyZiHrEFxe66Ygd2d3e3ohjYgYEdeAQLUVTsLkAp2fcPdLv3odmgiN97n/2cy8ysNbM4HD4ml5ZCoVAghBBCCAC0f3cDhBBCiOREAqMQQgihQgKjEEIIoUICoxBCCKFCAqMQQgihQgKjEEIIoUI3KW9eO0fdpLy9EL/Mubfev7sJQmgsKOhxkt079O2DBF+rlyFfIrZEc0kaGIUQQvwlwr/97hYkGgmMQgghNKcI/90tSDQSGIUQQmguXAKjEEIIoaRIQT1GWZUqhBBCqJAeoxBCCM3JUKoQQgihIomGUkNDQxk5ciTPnj0jJCSEXr16kSVLFnr27EmePHkAsLOzo0GDBjg7O7Nt2zZ0dXXp1asXNWvWJCgoiCFDhvDu3TsMDQ2ZPn066dKli7FOCYxCCCE0l0TbNVxdXTE1NWXmzJl8+PCBpk2b0qdPHzp16kTnzp2V5d68ecPGjRtxcXEhODgYe3t7qlSpwtatW7GwsKBfv34cOHCAJUuW4OTkFGOdEhiFEEJoLol6jPXq1cPGxkb5tY6ODjdv3uThw4e4ubmRO3duRo4ciZeXF6VKlUJfXx99fX1y5cqFt7c3Hh4edO3aFQArKyuWLFkSa50SGIUQQmguieYYDQ0NAQgICKB///4MGDCAkJAQWrZsSbFixVi6dCmLFy+mUKFCGBsbq10XEBBAQECAMt3Q0BB/f/9Y65RVqUIIITSmUIQn+BObFy9e0L59e5o0aULjxo2xtramWLFiAFhbW3P79m2MjIwIDAxUXhMYGIixsbFaemBgICYmJrHWJ4FRCCFEsvX27Vs6d+7MkCFDaNGiBQBdunTBy8sLgAsXLlC0aFEsLS3x8PAgODgYf39/fH19sbCwoHTp0pw+fRoAd3d3ypQpE2udWgqFQpFUDySHiIuUQg4RFylBUh4iHnzvfIKvNTCvHG3epEmTOHToEPny/TxofMCAAcycORM9PT0yZMjAxIkTMTIywtnZme3bt6NQKOjRowc2NjZ8/fqVYcOG8ebNG/T09Jg9ezYZM2aMsT0SGIWIAwmMIiVI0sDoczbB1xpYVE3ElmhOFt8IIYTQnLxdQwghhFCRgs5KlcAohBBCcynoSDhZlSqEEEKokB6jEEIIzclQqhBCCKEiBQ2lSmAUQgihMYVCVqUKIYQQP8lQqhBCCKFChlKFEEIIFSmoxyjbNYQQQggV0mMUQgihOTkSTgghhFCRgoZSJTAKIYTQnCy+EUIIIVRIj1EIIYRQIT1GIYQQQkUKCoyyXUMIIYRQIT1GIYQQGpOzUoUQQghVKWgoVQKjEEIIzcmqVCGEEEKF9BiFEEIIFdJjFEIIIVSkoB6jbNcQQgghVEiPUQghhOZkKFUIIYRQkYKGUiUwCiGE0JwExr9D+4EOdBjoEKeyL5+8pG2l9kncorgbOmcwNq3qcsT5KDMGzoqyTP4i+VhxdFmMZcSfzcGhBStXzom1XMuWXdm372iC7j148HgWLVqd0CYmWFDQ40hp4eHhBAeH8Pz5S9zczjB9+kKePXv5y9v2V5Kh1L+D5wVP1v/nd4pNS2uy5MyCy6pdBHwOVKYHfAr4xa2LG5tWdTm+y41rZ//93U0Rv5G7+wXc3S9Gm3/3ru8vbE3i+fjxE4sWrVF+raWlRerUqahQoTTduzvQuLENVas2kuD4K0iP8e/gecELzwteamklKll+D4y7efX01W9qWfw4TvsfXep0JyQo5Hc3Rfwm7u4XmTRp7u9uRqL79OlztM+1aNFUunZti5OTI716DfvFLfsLpaAeo2zXSOHu3bhHtjzZ4jwkLERKMWfOMgBsbGr95paIP40ExkRUopIlbk+P0tihEaMWjeDg/X04e2ylaNkiDJ0zGLenR8lfJF+k69yeHmX5kaWR0hu1a8iyQ4s5eH8fe266MHHNeAoUzR+vNq2YvIqP7z7SolvzKOuOTlzr1tbRxq5PG9a7r+Hg/X2sObGSeq1taNvfHrenR8mcI3O82it+v/TpzZgyZSTXr7vx/v1d3r+/y7Vrxxk6tA86OjoxXmtomIYZM8bg6XmCjx99ePz4Gtu2LadkyWKRymbOnJH58ydx//4lPn26h7f3WSZNGoGRkWGiPMfz5y+/P4+pWnqBAnlZu3Yefn4efP58n9u3zzB58khMTIzVyuno6DBq1ACuXDnC+/d3ef7ci337NlKzZpVEaV+KEx6e8E8yI4ExCbR3bEfBkhbsWbuXezfvc/9m/Odvhs0bguO0/6Grp8e+jQc4feAMlhWKs2DvPEpWLhnn+3x6/4ml45ejq6fLoJkD0daO/V95fOoes8yJriM6ExwUjOv6fTx98IwhswdRr1XdeD6xSA5MTIw5c8aVfv26cOfOPRYvXsP27XvIkiUTEyYMY9Kk4TFev3nzEvr378r9+49YuHANR46cxMamJm5uOzE3//mHWc6c2Th3bh/durXj339vsGDBanx8HjB4cC+OHXMmTZrUGj9L/vx5AHj+/OeUR7lyJbl48SCtWjXh0qVrLF26ntev3zJoUE/c3fdgZpZWWXbu3AmMHj2QDx8+smTJOlxc9lOuXEn27duIlVVFjduX4ijCE/5JZmSOMQmkNkpDd5tefHjzIUHXWzWsRt0W1hzffYLpA2YQ/i3iB2from0sPbiI4fOH0K5yB8JCw+J0v+O73KjTrDblapSlWZem7Fzpkih1V2tQlWr1q3L28Dkm9JzEt7CI97E16dCY/pP7JejZRdKwsqqIk5NjlHkbN+7Az+8pAN27O5AvX2569RrK2rXblGUmT57HzZunad3alhEjJkd5nyJFLKhXrxabNu2ka9eByvSDB93YunUZnTq1YeTIKQAsWDCZbNmy0Lx5Fw4dclOW7d27E3PmjGfUKEdGjZqS4OfV0tJSPu+ePYcA0NbWZs2aeRgY6GNr25Fjx04ry0+cOJwhQ3ozdeooevYcirGxEV262HPmzEXq1m2tLLd27TbOn99Pjx4dYlzM9FdKhj2/hJLAmARuXr6Z4KAIUN+uHgBLxi5VBiaI2BLiumE/7f5nT5lqpbl04nKc7zlvxAJWua2g4+D2nDl0NtqFQ/Gpu24LawCWTVyhDIoArhv2Y9vJllwFcsb9oUWSsrKqhJVVpSjz3N0vKAPjsWOn+fjxExs37lQr8/TpCx4+fIK5ed5o6/gxGlGokDlmZmn58OETAK6uRyhUqAqPHz8DIEuWTNjY1OTw4ZNqQRFg6dJ1ODp2p337lnEKjGnTmqgFfC0tLUxN01KrVlUKFzbHx8eXadMWAlCpUlnMzfOxebOLWlAEmDhxDnZ2TWnTxpb+/Z3Q1tZGS0uLnDmzkSNHVp4+fQHAtWteFC5clSdPnsfatr+OBEYRE01Xq1oUNyc4KJgmHf+JlPcj2OQvmj9egfHlk5esn7WenmN6MGBaf0a0G6Vx3QVLWPDp/Sde+L1QK6dQKLjtcVsCYzIyadLcOK1K9fS8hafnLQwN01ChQmny58+NuXk+ypQpQYECedDVjf5Xxs2b3ly4cJVKlcri63sZd/cLHDlyioMHj/Po0RNluZIli6GtrU26dKZR9mJDQkLJmTM72bJlVhsGjYqpaVq1e4SHhxMQEMiDB35Mm7aAefNW8unTZwAsLYsAcPbspSjqDMHDw5MmTepRsGB+bty4w86d+2jVqgm3b5/hwoWr35/FDW/vezF/E/9WCsXvbkGikcCYBIKDgjW63sjECF093RhXkpqYGkebFx2XVbupZVuL8jXKUbtpLR7dfaRR3WnTpeWJ79Moy7x79S7e7RO/n4GBARMmDKVr17YYGqYBInqL585d4s2b92TLFvNiqkaN2jFoUE/s7JpRr14t6tWrxdy5E3BzO0Pv3sPw83uKqakJABUrlqFixTLR3svMzDTWwOjn94SCBeO2GMbExAiAT5/8o8x/8SKirh/zm507O+Lh4UX79q2oXr0y1atXZsqUkVy96knv3sPw8rodp3r/GtJjFPGl+P7XlJaWllq6QSqDSGW/Bn7lS+BX7Cu0S9Q2hIeHM3vIHJYcWESvsT2Z3CfyUFV86v4S8IU0xmmizIsuXSRv06c70bNnB1xcDrB8+Xpu3LijHBK9ft0t1sAYGPiFCRPmMGHCHAoUyEudOla0aWNL7drV2LhxMVZWTQgIiDgYY8qUeUyYEPupPInF3z+i3qxZo36GHwtv3r2LmAYJCwtj/vyVzJ+/kpw5s1G7djWaN2+EtXV1du1aS6FCVQgLi9s8v/izyKrUX+THQpnUhuqr7bLlyRqp7APvh2TMmgGzjGaR8irWrkCnIR3JVzjuWy9U3b/ly86VLphlMKWHUzeN6vbxukfGrBlIlyldpLKFSxVKUPvE79W6tS2vXr2hbdteuLtfVAbFVKkMyJUrR4zXFi9emClTRlK+fCkA7t9/yLJl66lVqzn37j2gfPlS6OnpcfOmNwClS1tGeZ/RowcyeHBv9PT0EvHJwMvrFgBVqpSPlKelpUXlyuXw9w/g8eNn5MmTkwkThlK/fm0Anjx5zrp122nc2IGTJ8+RI0dW8uSRqQI1sl1DxNfj+xFzLBXrVFCmaWlp0baffaSyR5yPoq2tTf9JfdHV+9mpT5cpHf+b2h+7vq35+uVrgtuyfvZGnj96jnlxc43qPux8BG1tbXo4dVPbBlKnWW0KlZTA+CcKCgomVSoDTE1/blvQ1tZm9uzxyiHG6AKWgYE+Awf2ZMSI/mrpJibGmJqm5cWL14SGhvLo0RPOnLlIvXq1aNq0gVpZe/tmjBo1gLp1qxMaGpqoz3bu3BXu33+IrW09bGxqquWNGTOQnDmz4+Kyn5CQEL5+DWLQoF6MHTsIfX19ZTk9PT2yZMlEUFAQr169SdT2/fFku4aIrxN7TtBpSAda9WxJttzZePH4BWWsymCc1ijSYp0jzkepXLcSVg2rkbdQHq6c9kBHR4caja1Imy4tK6esirTgJT6Cg4KZN2IBM7ZOi5QXn7pP7j2FdfM61GlWm9wWufE870m2PNmoWKcCH999xDS9qdrKVpH8bd26i4EDe3L+/H5cXY+gq6uLtbUVBQsW4PXrt2TKlIH06c14+fJ1pGuvXvVk166DNGvWgIsXD3Lq1Hn09PT455+6ZMyYnh49BivL9u49HDe3nWzduozDh09w65YPFhb5aNCgNu/efaB/f6dEfzaFQkHXrgPZt28ju3at4cCB4zx44Kec67xz555yK8qrV29YuHA1AwZ059q1Yxw6dILw8HDq1q1B4cLmTJkyD3//5Hk+8m+TDHt+CSU9xl/kw9uPDGw5mH/PXqdczXI0sKvP43uP6d/UEf8oDiAf330ii8YsIehrMA3s6lHzn+r4+TxmTJdxbFvirHF7PM5c4+iOY1Hmxafucd0nsHnBFkxMjfmnQ2Oy5c7KtP/N4N9z1wEI/hqkcVvFrzN27EwmTJhNeHg4PXq0p0kTG/z8ntKoUVumT18EQL16NaO9vnPn/+HkNA1dXV26dLHHwaEFDx740axZZ9av//mzc+/eAypXbsjq1VsoVqwwfft2wtKyMFu27KJq1cZJtvLz4kUPqlRpzM6d+6hYsQw9e7YnXTpTpk6dT9WqjZVDxwAjR06hb9+RfP4cgINDC7p0scffP4CuXQf+0rnRP4ZCkfBPMqOlUCRdq2rnkNNPUrKMWTMS6B/Il4AvkfLm7JyFhaUFjSwib/v4E5176/27myCExqJ6VVdi+bp2aIKvTd1pRiK2RHPSYxQJ1qZPK/Z578GyYnG19MKlC1GsXFE8L3j+ppYJIX65FLT4RuYYRYId3naEBnb1mbJ+EmcOnuXty7dkyZmFKvUq8zXgK8snrvjdTRRCiHiTwCgS7N7N+/T753/Y9W1DySolMU2flk/vP3HK9TQb52/WaIGQEOIPkwxXlyaUBEahkfu3fJnYK+pDpYUQfw9FePJbRJNQEhiFEEJoLhnOFSaUBEYhhBCak6FUkRQ6DelIu/9FPgkHIjbTT/p+tmkDu3oMmjkwynK3r92h3z//U36traNN656tsGllTebsmXn76h2n959m6+LtBH4OTPyHEOK7rFkzc/26GxMnzmXRotVqeWnSpGbQoF60aNGIXLly8OLFK3bscGX69EV8+c+pTnp6ejg6dsfOrhl58+YkODiEq1c9mT59obwTMTmRoVSRFPIVzktIUAhbl2yPlKf6Joy8388q3bp4GyHB6sdmvX3x85gqLS0txq8cS+W6lXjx+CUHth7ENJ0prXq0pGLtigxqNYRP7z8hRGIzNEzDtm3LSZvWJFKejo4Oe/asw8qqEqdOnePAgeNYWhZh2LB+1KlTnVq1mhMcHPGGGi0tLXbtWoO1dXVu3LjDihWbMDU1oVmzhhw6tBUHh77s2nXgVz+eiIoMpYqkkK9wXvzu+bFhzsZYy33+8JlVU9fEWK5uizpUrluJW1dvMaztSL4GRvwlXt6lHFM3TKaHUzdmDJyVaO0XAiBXruxs27aC0qWLR5nfsWNrrKwqMX/+SoYNm6hMnzhxGEOG9KFjx9YsX74BgBYtIt5msXv3Qdq168O3bxEvxJ41aynnzu1j3ryJ7N9/jJCQkKR/MPHXkA3+yUQaozRkyZmFB3cexlo2X6G8PPCOvVyNJjUAWDphuTIoAlw+cYWrpz2o06w2adOljeZqIeKvb98uXL16FEvLwpw8eS7KMvnz5+HNm3fMmrVELX379r0AVKxYWpnWpEl9ACZOnKMMigA+Pr7s3LmPTJkyUKZM1G/pEL9YEm3wDw0NZciQIdjb29OiRQvc3Nzw8/PDzs4Oe3t7xo4dS/j3ezg7O9OsWTNatWrFyZMnAQgKCqJfv37Y29vTrVs33r9/H+ujSGBMJvIVzgsQa2DMkDUDJmYmcQqgWXNmISw0DB+vyOdOPrjzAB1dHYqULpywBgsRhX79OvP48TPq1GnJli0uUZYZOXIKOXOW4s0b9ZdZFyxYAIBXr94q01xc9jN16nx8fB5Euk9wcEQv8ccLlcVvlkRnpbq6umJqasqWLVtYuXIlEydOZOrUqQwYMIAtW7agUChwc3PjzZs3bNy4kW3btrF69WrmzJlDSEgIW7duxcLCgi1btmBra8uSJUtirA9kKDXZ+PGOw7TpTJixZRoWlhGvhPr33HVWT1/L0wdPv5eLCKC6ujqMXzWWomWLYpBKn1tXb7N21nruXr+rvGdoSCha2lro6OrwLeybWn2GJoYAZMqRKcmfTfw9+vQZwYkTZwkPD8fcPG+crjEzS0vdujWYPXs8Hz58YsWKn1MJu3cfZPfug5Gu0dPTUx5mnlQHjot4SqI5xnr16mFjY6P8WkdHh1u3blG+fMR7Na2srDh37hza2tqUKlUKfX199PX1yZUrF97e3nh4eNC1a1dl2bgERukxJhM/Al6rni0JDAjk4NZD3LnujVXDaizet4D8RfJ9Lxfxz3/aN0Y/lT5HnI/gceYapauWYp7LbMpWL6O8511PH3R0dKhiU1mtLj0DPcpUixiuMjQ2/BWPJ/4Sx4+7K4e14qJjx9a8eHGD9esXkiqVAc2adeLBA79Yrxs6tA958uTi8OETPH0qJywlC+GKhH9iYGhoiJGREQEBAfTv358BAwagUCjQ0tJS5vv7+xMQEICxsbHadQEBAWrpP8rGRnqMyUT4t3BePnnJjIGz8LzgpUyv3bQWIxcOZ8jsQfSs3wdtLS1ePnnJmhnrcNt9QlnOsmJxZm2bwZDZg2lXpT2hwaHsXrMH6+Z16D+5LwCXT1zGLIMZ3Z26KecWf/xwCfE7vHv3gXnzVpAlS0Zsbeuzb99G2rTpwfHj7tFe07Ztc0aNGsDHj58YMGD0L2ytiFES7mN88eIFffr0wd7ensaNGzNz5kxlXmBgICYmJhgZGREYGKiWbmxsrJb+o2xs4txjjM9fgSL+Fjgtom2l9mpBEcBt9wk8L3phXtycHPlysGXRNtpWaq8WFAG8Lt7AbfcJMmRJT4mKEYsRfG8/YNqAGegb6OO0eCSud/aw/sxaMmXLyOrpawEI/hr8ax5QiCjs23eU4cMn0bHj/6hRoxm6ujqsWTOPNGlSR1m+c2c7VqyYRXBwCK1bd+fRoye/uMUiWknUY3z79i2dO3dmyJAhtGjRAoAiRYpw6dIlANzd3SlbtiyWlpZ4eHgQHByMv78/vr6+WFhYULp0aU6fPq0sW6ZMmWjr+iHGHuOTJ0+YOnUqN2/eRFdXl/DwcCwsLBgxYgR588Zt/kBo7t6N+5SoaEnWXFmUc41Rlrt5j7otrcmSM4sy7eTeU3he8KKSdUWM0xrx0PsRV05dpbFDQwA+vP2Q5O0XIi6uX7/Jli276dzZjooVy3DixFm1fCcnR5ycHAkICKRVq26cPn3hN7VU/ErLli3j8+fPLFmyRDk/OGrUKCZNmsScOXPIly8fNjY26Ojo4ODggL29PQqFAkdHRwwMDLCzs2PYsGHY2dmhp6fH7NmzY60zxsA4atQoBg0aRIkSJZRp169fZ8SIEWzbtk3DxxU/aOtoY16sAFra2nj/G/mFuAap9AEICQ7BvFgBUhmm5salG5HK6acyACA0WH1P1/vX7zmwWX0Bg4WlBQB+PrHP5wiRmKpWLY+paVr27z8WKe/x44g//NKnN1NLX7hwCt26tePduw/Y2nbgypXrv6KpIh4USTSq6OTkhJOTU6T0TZs2RUpr1aoVrVq1UktLnTo1CxYsiFedMQ6lhoSEqAVFgJIlS8arAhE7HR0d5u+ey9SNk9DWjvyvpGjZIoSFhnH/li/jV49jtvMMTMwij5MXL1cUgLvft2c07WzL7ps7lStcf9DT16NC7Qq8e/UO39uRl8ELkZSWLZvJ1q3LMDOLvIe2ePEiAGoLcKZPH023bu14+vQFtWu3kKCYXCXRUOrvEGNgLFiwICNGjODgwYOcOXOGw4cPM2LECAoWLPir2vdXCA0J5cLxi5iYmtCmT2u1vJY9WpCvcD5O7DlJ4OdA3Pe7o6OjQ5fhndXKWTWsRsU6FfG86KU8Ps73ti8mpiY0dmikVrbfpD6YZTDFedlOFLHsIRIisbm47EdPT48JE4appderV4umTetz48YdPDwi5tobNrTmf//rxtu376lbt5VszUjOFOEJ/yQzMQ6ljhs3juPHj+Ph4UFAQABGRkbUrFkTa2vrX9W+v8ayCSsoWqYIXYZ1omSlEvje9sXC0pySlUvi5+PH0vHLANg4fzPla5ajUdsG5Cucl5uXb5Izf04q1C7P25fvmKlyxJvXxRu4HzxDA7v6ZMyaEd/bvhQtW5Ti5Ytx6cRl9qzb+7seV/zFZs5cQv36tenWrR3FihXiwoWrFCiQl0aNrHn//iMdOvRTlh03bjAAN27cwd6+WZT3c3Z2xcfH95e0XcQgGfb8EkpLkYRdhto56ibVrVOkDFnS03FwB8rXLIeJmQnvXr3D/cAZNs3fTKD/F2U5QxND2js6UK1+FdJlSsen95+5dOIS62Zt4P1r9eOO9Az0aNvPjpr/1CBD1gy8fPySIzuOsXvNHkJDQv/bBBGNc28jz/2KmDk4tGDlyjkMHjw+0ts1jIwMGTVqAM2aNSBr1sy8e/eRI0dOMHnyPJ48eQ6AiYkxr1/firWeli27sm/f0SR5hpQmKOhxkt07cJxdgq81HLc1EVuiOQmMQsSBBEaREiRpYBzTJsHXGk5IXos55eQbIYQQQoWcfCOEEEJzyXARTUJJYBRCCKG5FLT4RgKjEEIIjSXVBv/fQQKjEEIIzUmPUQghhFCRggKjrEpNZOkzp2Pv7d0069I0yvxyNcoye8dMXO/sZpfXDqZumkzBEhYJrm/s8tG4PT1K5hyZI+WlTZeW/03px/arWzlwz5XlR5bS2KFRlK+aamBfn62XN7P39m4mrB5HhizpI5XJVSAnR/0OUalOxQS3V/yZ0qUzZfbs8dy+fYYPH3z49183HB17oKOjE6frixUrxI4dq3jx4gYfP/pw5coRunSxj7Jsnjw5WbduPk+e/Mvbt3c4fXoPLVo0jrLs8OH9ePz4Gs+eebJq1RzSpo18VGKdOlZ8+fKI4sULx/2BRfyloJNvJDAmolRpUjFu5ViMTKJ++W8D+/pM2zSFHHmzc3j7Ec4fvUCJipbM2zUnQcGxav0qWDWsFmWeaXpTFu1bwD/tG/P6+Wv2bzxAwOcABkztz6hFI9TKmhc3Z9AMR96+fMvh7UcoXqEY41eNjXTPTkM6ctfThwvHL8a7reLPZWRkyIkTLvTp04nbt31YunQdnz59ZurUUTg7r4z1+uLFC3Pq1G7q16/F0aMnWbFiE0ZGhixePI1Jk9R/Fi0s8nPu3H7atGnKjRt3WL16C7q6umzatJjJk0eqlW3SpB7jxg3B0/MWO3a40rx5I5YsmR6p/nHjhuDicoAbN+5o9o0Qfw0ZSk0kmbJnYvzKsZEO7FbmZ8tIn/G98PPxY0DzQXz+8BmA/ZsPsmDPXLqN7Mrg1kPjXJ9RWiP+N7lftPndR3UlW+6s7Fqzh8Vjlqilt+7ViiunrnBkR8TbDeq3tuHzx88MajWEkKAQvP/1xmnJSMyLm3PvRsTZlObFClC1fhWGtx0ZZX0i5Ro6tA+FCpkzcOBYlixZq0xfv34BrVvbUq9eLQ4fPhHt9ePGDcHIyFDthJrx42dx4cIBHB27s2rVJuV7FRctmkL69GYMHz6JefNWAKCtrc26dfMZNKgnR46cwN094g+zjh1bc+fOPRo3dgDg6dMXTJo0nH79THn//iMA//xjQ8mSRenceUBif1vEf8lQqlDVrEtTVh1fTv4i+bh29t8oy9S3q0eq1KlYNGaJMigCeP/rzfalO/C9Fb+zHvuM64Wuvi63PW5HytPW0aZag6p8/vCZVVPUj+JaO2s9gf6BNO/289zJLLmy8PTBM0KCIl5Xdf97W7Lm+vlex05DO+J16QYeZ67Fq53iz5c7dw6ePHnG8uUb1NKdnfcBULFi6RivL1PGkvfvP6od2xYY+AVnZ1d0dHQoW7YkAJkzZ8TKqhJ37txTBkWIeEn6sGGTAOjXr6syPU+eXGq9QE/PW8p0AC0tLcaMGcTWrbvlLNVfQBGuSPAnuZEeYyJo3rUpr569Zu6w+eTMl53SVUtFKlO+Zjk+f/zMv+euR8pbPW1NvOorV6MsdVtaM8NxJlVsqkTKN01vShqjNFy/4ElwULBaXmhwKE8fPMO8eAHSGKXhS8AX/D8FkClbJmWZNMZpAAj8HAhEvPaqQq3y/K+pY7zaKVKGDh36R5lesGB+AF69ehvj9e/ff8TCIh+mpmn5+PGTMj1btog/vN6+fQdEzC0CXL16PdI9Xrx4xdu376lcuZwy7ePHTxgZ/Zy2MDExBuDTp4g/PFu3boKFRT5atOgSY/tEIkmGAS6hpMeYCOYOn0+Pur2i7L39kNs8N0/uPyFdJjOGzR2Ci6cz+31cmbZpCvmL5ItzXakNU+M4fQAe7teUQ6H/9eNwcH19vSjzDU0M0dbWJlP2iGB459od8hbKQ5V6VUhjlIYWXZvx9UuQsufYeVgnLp+6ws0rsR/oLFK+jBnT0727A6NHD+Tx46ds3bo7xvIrV25CV1eX9esXkC9fboyMDOnQoRUODi24du2Gcmg0+PsLtg0M9CPdQ0tLCyOjNKRPb4ahYcQfbpcuXaNmzSpUrFiG9OnN6NHDgefPX/Ho0RN0dHRwcnJk/Xpn5TCtSGLh4Qn/JDPSY0wEV097xJhvaGJIasPU6Bvos3j/QoK+BHFiz0nSZUpHtfpVmb97LgNbDsbHK/Z3zXUf1RUTMxPmDp8XbRn/j/4893tB/qL5yZIzCy+fvFTm5bbIrRwiNfzeMzyw5SB1W1gz4fuCm2/fvrF47FI+vf9EmWqlKVmpBL0b9lXeQ0tLS97j+JcaO3YQI0b8D4CXL1/TsGE7tV5gVJYuXUdYWBizZ4/j9u0zyvTjx91xcOhD+PdfjHfu3OPLl69Ur14ZExNjPn/2V5a1tq5OqlSpAEib1pjAwC/Mnr2Uf/6x4dSpiMAcFBRE+/b9+PbtG50725EjR1amTfv55nb5uU1i0mMU8ZE6TcR/0ObFzXl8/wndbXqxeOxSJvaazNhu40ltmJqB0wfEeh/LisVp1K4h62et58XjlzGW3bliJwapDJi4djxFyxYhVZpUFCtXlLHLRyvnEn9s2wgNDqW/7QAm9prMknHL6NOwH3vXuQIRc4tnD5/jrqcPBUsWZNXxFRz1O8Sak6soUclSg++K+BM9fvyMOXOWsWfPITJmTI+b205KliwW4zXly5di6NA+hISEsmnTThYtWsOdO/eoVasqY8cOVpYLDg5m2bL1ZM6ckV271mBpWQRDwzTY2NRk6dIZBAREDO3/+Ll98+YdFSrUp2vXgTg6jqFcuXq4uh5BX1+fESP6s2LFJp49e4mNTU28vc8SGPiQs2f3UaxYoaT7Bv3NwhUJ/yQz0mP8BcJV/sUvm7BCGZgALhy7yPXz1ylZuSTZ82bj2cPnUd5DP5U+g2YOxMfrHi6rYh66Ati7fh/Z82anaWdbFuyZp0w/vssNzwue/NO+McFff84/hoWGcWrfabV7VK5biYIlLOhetyc6ujqMWzGGZw+fscxhOfXb1GP8yrG0q9KBgE8Bcf1WiD/c2rU/Xw9Ur14tdu1aw+rVcylTJuqXlxsbG7F79zq0tbWpWLEB9+8/BEBPT4916xbQs2cHbt/2YcWKjQCMHTuTnDmz07JlYy5fPgxELL6ZM2c5hQub07BhHb58+aq8v79/AJs27VSrs1u3dpiZmTJz5mLSpzdjy5alHDzoRt++Ixk6tA/bt6+gRIlahIWFJer3RqQcEhh/gR+LWEJDQnl091Gk/Pu3HlCyckmy5Y4+MHYa3IEsOTLTq/sE5dBTbJaMW8ahbYcpXbU0WlpaeF3ywsfrHmOWOQHw4e2HGK/vNKQDp1xP89D7EZWsK5IpW0ZGth/FQ+9HeF+/yy7PHdRuWkvZuxR/l8OHT3Dy5Dlq165Gvny5efDAL1KZRo2sSZ/ejMmT5ymDIkBoaCgDBjjRvHlDHBxaKgNjaGgoDg59WLRoNZUrlyM4OITjx93x8fHl7Nl9BAcH8+FD9EO3adKkZsiQ3ixduo43b97Ru3cnDAz06ddvJB8/fuLly9dcvXoUa+vqHDrklvjflL9YShqmlsD4CwQHBfP25VvMMpqhpa0F/4lrunoRp4cEfQ2O4uoIVg2roauny8pjy6PM33Ix4hfLf18O/dD7EQ+9H6mlWVhaEPApgLcv30VbX80mNchtnpvx3ScCkCNvdgCePnwGQMCnAD69/0S23NmivYf48+no6FC9eiW0tLRwczsTKf/x46cAZMiQLsrAmCNHxM+Ht3fk+fM3b97x5s07cuaM/DN06dI1Ll36uTXIwMCAokULcvduzNsu+vTpTOrUqZgzZxkABQrk4e3b98p50Hv3IoJzvny5Y7yPSIBkOCSaUBIYf5Ebl25Ss0kNSlS0jLTX0aK4OWGhYfj5RP7F8oPLqt0YpTWKlF7znxrkKpATl1W7CPjeMwUYtWgElhWLY1e+nVoPs0DR/GTNlSXSsKkqbR1tOg5qz9Gdx5SBUEc3Injr6OgQyvdVrwb6kIL+ShRRc3FZg79/AHnylI00WlG8eBHCw8OjXfn5+vUbAMzNI6+8NjVNS/r0Zmp7EU+ccEFfX5+qVdWPgLOxqUHq1Kk4dsw92naamBjj6NiD+fNXKnuVurq66Or+/DWXKpUBkLJ6N8lGCgqMsvjmF9m/+SAQsao0tWFqZXqNxtUpUqYIF45fVNv4/1+7Vu9mw5yNkT5P7kf8QnJZFZH/w2PfJ2TIkoFatjWVaYbGaRg0cyAA25Y4R1uXTcu6ZM6RmY3zNivTnvhG9AyKlI44bzJn/pwYpTXiyYOncf4eiD/Pt2/f2Lv3EJkyZWDgwJ5qed26taNs2RIcOnSC16+j3st48KAbgYFf6NWrI3nz5lKma2trM2PGaLS1tXF2/jkUf//+Q8qWLUGVKuWVaTlyZGXGjNF8/RrEokXqB1aocnTsTnh4OAsWrFKm+fg8IEOGdOTPnweIWAgEcO/eg7h/E0ScyAZ/EW/Xz19n1+qIw8VXu63gzMGzZMiagWoNqvL+9XuWjFumVr5Zl6YYpTXCZdUu5RxlfLis3IVNy7oMnjWQMlal+fj2I1XrVSFbnmysnbleedTbf+nq6eIwoC0Htx7m1dNXyvQrp6/y+vkbhs8fysm9p6hUtyKf3n/ixJ6T8W6b+LOMHDmFqlUrMGnScKpXr8SNG3coWbIYtWpV5eHDx/Tt+/O8UyeniEMgJk2aC0QMlzo6jmbp0hlcunSI3bsP8vHjZ2rUqEyJEkVxd7/AwoU/g93EiXNo3NiGvXvXs23bHoKCgmnRohGZMmWge/fBvHjxiqhkyJCOvn27MHXqAuXqVYCdO/cxduwg9u5dz8GDbrRu3QRv73tRDgsLDSXDAJdQ0mP8hRaPXcoMx5l8fPeJxu0bUaJSCU7sOUmfxv15/ey1WtnmXZvSYaADRiaRh0/j4kvAF/7XdACn9p2mdNVSNGzbgPdvPjC223g2zd8c7XWN2jXENIMpmxeolwkJCmFk+1G8evqKfzo05kvAV5w6juFLwJcEtU/8OZ4/f0WVKo1ZvXoLRYsWol+/LhQokIcFC1ZRpUojtWDl5OSoDI4/bNiwgwYN7Ll06RpNmtSjZ8/2GBjoM27cTBo1ciAk5Ocq7SdPnmNl1YQTJ87yzz822Ns34/ZtH+rXt4u0+lTV4MG9CQj4wtKl69TSX758TbNmnQgODqFHDwfu3XtAy5bd4ryATcRDuAafZEZLkYSD7f9dCCLEn+rcW+/f3QQhNBYU9DjJ7v3JoXaCr027MXmtEJahVCGEEBpLjnOFCSWBUQghhOYkMAohhBAqkuFcYUJJYBRCCKExGUoVQgghVEmPUQghhPgpJfUYZR+jEEIIoUJ6jEIIITQnQ6lCCCHETwoJjEIIIYQKCYxCCCHET9JjFEIIIVRJYBRCCCF+Skk9RtmuIYQQQqiQHqMQQgiNpaQeowRGIYQQGpPAKIQQQqhSaP3uFiQaCYxCCCE0Jj1GIYQQQoUiXHqMQgghhFJK6jHKdg0hhBBChfQYhRBCaEwhi2+EEEKIn1LSUKoERiGEEBqTxTdCCCGECoXid7cg8cjiGyGEEBpThGsl+BMXnp6eODg4AHDr1i2qVauGg4MDDg4OHDx4EABnZ2eaNWtGq1atOHnyJABBQUH069cPe3t7unXrxvv372OtS3qMQgghNJaUQ6krV67E1dWV1KlTA3D79m06depE586dlWXevHnDxo0bcXFxITg4GHt7e6pUqcLWrVuxsLCgX79+HDhwgCVLluDk5BRjfdJjFEIIkazlypWLhQsXKr++efMmp06dom3btowcOZKAgAC8vLwoVaoU+vr6GBsbkytXLry9vfHw8KBatWoAWFlZceHChVjrk8AohBBCYwpFwj+xsbGxQVf35wCnpaUlQ4cOZfPmzeTMmZPFixcTEBCAsbGxsoyhoSEBAQFq6YaGhvj7+8danwRGIYQQGkvqOUZV1tbWFCtWTPn/b9++jZGREYGBgcoygYGBGBsbq6UHBgZiYmIS6/0lMAohhNCYQqGV4E98denSBS8vLwAuXLhA0aJFsbS0xMPDg+DgYPz9/fH19cXCwoLSpUtz+vRpANzd3SlTpkys95fFN0IIITT2Kzf4jxs3jokTJ6Knp0eGDBmYOHEiRkZGODg4YG9vj0KhwNHREQMDA+zs7Bg2bBh2dnbo6ekxe/bsWO+vpVAk3e6T2jnqJtWthfilzr31/t1NEEJjQUGPk+zePoXrJfhaizuHE7ElmpMeoxBCCI2lpLNSZY5RCCGEUCE9RiGEEBqTs1KFEEIIFSnprFQJjEIIITQmPUYhhBBCRXgKWnwjgVEIIYTGUtKqVAmMQgghNJaS5hhlu4YQQgihQnqMQgghNCZzjEIIIYQKmWMUQgghVKSkOUYJjEIIITQmQ6lxdPr1raS8vRC/zNfnZ353E4RI1mQoVQghhFCRknqMsl1DCCGEUCE9RiGEEBpLQWtvJDAKIYTQXEoaSpXAKIQQQmOy+EYIIYRQEf67G5CIJDAKIYTQmALpMQohhBBK4Slo9Y1s1xBCCCFUSI9RCCGExsJlKFUIIYT4SeYYhRBCCBWyKlUIIYRQIT1GIYQQQoX0GIUQQggVKSkwynYNIYQQQoX0GIUQQmhM5hiFEEIIFeEpJy5KYBRCCKE52eAvhBBCqEhBR6VKYBRCCKG5lLQqVQKjEEIIjYVrpZyhVNmuIYQQQqiQHqMQQgiNyRyjEEIIoULmGIUQQggVso9RCCGEUCH7GIUQQggVKWmOUValCiGEECqkxyiEEEJjMscohBBCqJBVqUIIIYSKlDTHKIFRCCGExmQoVQghhFAhQ6lCCCGEipQUGGW7hhBCCKFCeoxCCCE0ppA5RiGEEOInGUoVQgghVIRr8IkLT09PHBwcAPDz88POzg57e3vGjh1LeHjEXZydnWnWrBmtWrXi5MmTAAQFBdGvXz/s7e3p1q0b79+/j7UuCYxCCCE0ptDgE5uVK1fi5OREcHAwAFOnTmXAgAFs2bIFhUKBm5sbb968YePGjWzbto3Vq1czZ84cQkJC2Lp1KxYWFmzZsgVbW1uWLFkSa30SGIUQQmgsXCvhn9jkypWLhQsXKr++desW5cuXB8DKyorz58/j5eVFqVKl0NfXx9jYmFy5cuHt7Y2HhwfVqlVTlr1w4UKs9UlgFEIIobGkHEq1sbFBV/fnkhiFQoGWVkRENTQ0xN/fn4CAAIyNjZVlDA0NCQgIUEv/UTY2EhiFEEL8UbS1f4auwMBATExMMDIyIjAwUC3d2NhYLf1H2djIqtT/aO/QijWr58ZarlmLzri6HknQvQcOGsuChasS2sQECwt5BkAd65acOn0+yjKzZ43nf/27UrtOC067xz7kIP4ci1dvYumazXEqmy1LJo66rE/iFsXdqEmz2XvoeKR0XV1dTIwMKVywAO1aNqFapXK/oXUCfu2q1CJFinDp0iUqVKiAu7s7FStWxNLSknnz5hEcHExISAi+vr5YWFhQunRpTp8+jaWlJe7u7pQpUybW+0tgjMbp0+djDAx3797/ha1JXEuXzKBk6drKiWzxdyhXyhI6q6ftPXiM5y9f065lE4yNjZTpJkaGv7h1cdOkfh2yZc2s/DosLIyXr99y9MQZzl3yYNyw/9Hin3q/sYV/r195iPiwYcMYPXo0c+bMIV++fNjY2KCjo4ODgwP29vYoFAocHR0xMDDAzs6OYcOGYWdnh56eHrNnz471/hIYo3Ha/QITJs753c1IEubmeRnt5IjT6Gm/uyniFypf2pLypS3V0q7868Xzl69xaN2U7CoBJ7lq0sA60jMAtGxSnw69hzBr0Uoa1q1B6lSpfkPr/m5JfYh4jhw5cHZ2BiBv3rxs2rQpUplWrVrRqlUrtbTUqVOzYMGCeNUlc4x/GT+/p7x9+55BA3tSvHjh390cIRJFacuilCxemIDAL1zzvPW7m/NXSup9jL+SBMZEkD69GdOnOnHD6xSfP97n88f7eF4/wfBh/dDR0YnxWkPDNMyeOY6bN04T8NmX50892eG8klIli0UqmzlzRhYumMKjB1cJ9H/AvbsXmDplJEbxGPb68PETQ4ZNQE9Pj+VLZyhXdsUmPnXny5ebTRsX8+zJdT6+92Hf3g0UKlQA79tncTu2I85tFb/X5WteFKtSn2279zNk7DTK1GxCjX/sueZ1i1GTZlOsSn28fXwjXVesSn2ad+ijlqZQKNi++wAtO/WlTM0mVK7Xkr5Dx3HHJ/GmJDJlyADAp88/Vx2GhISwYv02/mnbnVI1GlOlfiv6DRvPzTs+ka4/d8mDLv2HY9WwDWVr2dLUoRcrN2wnNDQ00dqYkiXlPsZfTYZSNWRiYsz5s/vJlSs7+/YfxdX1MBkypKepbX0mTRxOOjNThg6fGO3127Yso3792uw/cAxX18NkzpyJVi0bU9e6BuUq1MPn+y+enDmz4X5qL9mzZ2H/gWN4e9+jhGVRhgzuQ+3aVtSo2ZQvX77Gqc0bN+6grV0z6tSxom+fzixctDrG8vGpO3/+PJw5vZf06c3Y63qYhw8e07CRNadP7kFbW4tnz17E8Tsrkoula7aQJnUq7Fv8w/2HfhS2yB/ve4ycNJt9h90okDc3rWwb8DUoiCMnztCuxyCWzBpPhTIlNW7nk2fPAciUMT0AwcEhdBswkmtetzDPn4fWtg15+/4DJ85c4MzFq8yZNJJa1SoBcM3zJn2HjcMsbVrq1bbCwMCAC1euMX/5Oh4/e87EEY4aty+lC0+WIS5hJDBGo7pVJcaMHhhl3voNzvj5PQWgZ4/25M+fh+49BrNm7VZlmYmT5uJ9+wxt2thGGxiLFi1I/fq12bBxB527DFCmHzh4HOdtK+jSyY5hIyYBsHjhNLJnz0LTZp04cPDn6ry+fTozb+5ExoweyPARk+P8fL37Duf6NTcmjB/Knr2HePLkebRl41P37FnjyJQpA63adGfXrgMAjB47g6OHt1G1aoU4t08kH1++fMFl3SIypE+XoOuPnDjDvsNuNLSuwWSnwejqRoyidHVoTesu/Rk5cRaHd6xFT08vwW10cz/P7bv3yZQhPSWKFgJg7ZadXPO6hW0Da8YN+5+y3lve9+jQewhOk+dw1GUdRoaGbHTeQ2hoGBuWziJHtiwAhIZ1wK7r/3A9dJyh/bpjnEwXJInEJ4ExGtWrV6Z69cpR5p0+fUEZGI8eO8WHD5/YsFF9iPDp0+c8ePgYC/N80dbxYy9O4cLmmJmZ8uHDRwD27j1MAYuKPH4csb0iS5ZM1KtXk0OHTqgFJoDFS9Yy0LEnHdq3jldgfPDAjwkTZzNtqhOLF07jH9v2UZaLT93p05tRv14tzpy5qAyKEDGcNWLkFM64741z+0TyUcqyaIKDIsCu/RHbmob9r4cyOAHkyJaF1k0bsmL9Ns5f+ZfqlcvHeq+9B49x5V8v5dchIaH4+D7k/GUPdHV0GD24rzLA7jl4jNSpDBgxoKdavUULmdOmWSPWbXXh+Knz2Da0Jjw8ordzzeuWMjDq6eqybPZEDAwMJCjGQXKcK0woCYzRmDBxdpxWpV6/fovr129haJiGihVKkz9/Hiws8lG2TEnMC+RVO63hv27cuMOFC1epVKksfg+vcvr0eQ4fOcn+A8d49OiJslzpUsXR1tYmfXqzKHuxISEh5MqVnWzZsvD8+cs4P+PceSto3dqWBg1q06rVPzg7u0YqE5+6ixUtiI6ODleuXI9U7tLlazJX84fSdLXq7bv3MNDXZ+uu/ZHyHvpF/Jx73/ONW2D8z15GfX090puZUbdmNdq3bkrxIgUBCAz8wtPnLyllWQRDwzSR7lO6RFHWbXXh7v0HALT4px4nzlxg5MRZLF+3laoVy1KtYlkqlCmhUU/2b5JyBlIlMGrMwMCAyROH061bO+V/gE+fvuDM2Yu8efOObN//+oxOvQZ2DBncG3u7ZtSvX5v69Wszf94kjh93p0evIfj5PSWtacRJDRUrlqFixeg3p6ZLZxqvwPjt2zd69BjM+XP7mTt7AseOuUcqE5+602eI6FW8fPU6Un54eDivX7+Nc9tE8mGgr6/R9f7+gYR9+xbj4QKfPwfE6V5rFk6PcrvGfwV8+QKAkWHUPb2M339Wg77v5a1WqRxrFk5j7RYXLl79l8079rJ5x17SmhjTu3Nb2rZsEqf2/c2kxyiUZs4YQ+9eHdnpsp+lS9fhdeOOckj0htepWANjYOAXxo2fxbjxszA3z4d1HSvsvy+M2bp5KZWrNiYgIOI4o0mT5zJu/KxEbf+1f2+wcOFqHB17MGP66Ei/oOJTd57cOYGIBUlRUd1ALv5sP1YzK/7TT/gaFBSpbJo0qUmTJjXHd234JW0DMEyTGoA3b99Fmf/j5zytyvFg5UpZUq6UJV++BnHN8yanz11m76HjTJ23jFw5ssmpOrFI6n2Mv5Js19CQXRtbXr16Qxu7Hpx2v6AMiqlSpSJ3rhwxXmtpWYTpU52oUL40APfuPWDJ0nVY1bDF594DypcvjZ6eHjdu3AGgTDR/KY8dM4ihQ/okeMhn7PiZPHz4mE4d21CtmvoCmfjUfe3fG4SHh1OubMlI5QoXNo82YIo/j55exN/UX76oB8InUaw6tsifh1ev3/L2XeT34J06d4kFK9bjfe9BorbPyNCQHNmy8OjxM95//29SlYfnTQAK5M0FwEbnPSxcERG406RORdWKZRk1qDdOg/t8Ly97I2MTjiLBn+RGAqOGgoKCSZXKAFPTtMo0bW1t5s4ZT5rvf7VGF7AM9PUZNKgXo0YOUEs3MTHGzDQtL168IjQ0lEePnuDufoH69WvTrFlDtbJt2zZntNNAbOrWSPAc3pcvX+nbbwQQMaeoKj51P3/+kuPH3alTx4r69Wopy+nr6zNtilOC2iaSp7y5IkYHTp+/pEwLDw9nxfptkco2aWCNQqFg8pwlaj+jb96+Z+KsRaza6Eya1Il/Uk2T+nUICg5m+oIVhIV9U6bfvnufLTtdMTE2osb3ldLnL3mwYsM2PG/eUbvH8xevgIizY0XMZB+jUNqyZReDBvXi0oWDuLoeQUdXh7p1a1CoYAFev35LpkwZSJ/ejJcvI8+7Xbl6HZddB2jerCFXLh/h1Mlz6Orp0uSfemTMmJ6u3X4udunZexinTuzCedsKDh1y49btu1hY5Kdhgzq8e/eBvv1HavQcR46eYvMWF9raN4+UF5+6Bwwcw7kzruzetZa9rkd4+vQF1tZWZMwQsbfs27eUNBPx92pQtwYLV21g7RYXnjx7QY5sWTl/+Rqf/QPImlk9iNg2sObk2YscO3WOpg69qFyhDN++fePIiTN8/PQZx16dyJUjW6K3sXPblpy7fI0DR0/i4/uQCqVL8O7DR064X0CBglkTRijnIHt3bcfla1507jecurWqkSlDeh48esypc5fIlycnjWxqxVKbSEmkx6ghpzHTGTd+JuHhCnr2bI9tk/r4PXpC/QZ2TJ0WcT6fau/pvzp07M/IUVPQ1dWha9e2dGjfigcP/LBt2pF167cry/n4+FK+Yn1WrtpE8eJF6Ne3C5bFi7BpswsVKzfgzp17Gj/LoMHjePs28nBXfOr28fHFqoYtBw+5UbtWVbp2sefBAz+sbSLOL/zyNW6HEIjkLUM6M9YunE6FMiU5e/EqLvsOky9PTjYum4XJf+aStbS0mDtpFMMH9CRVKgN27TvCYTd38ufJxfypo+nSrlU0tWjGwECfVfOn0LerA2GhYWzfc4BLHp7UqFqBzcvnKDf3AxQvXJB1S2ZQuXxpLntcZ8P2Xfj4PqRdyyZsWDIrSXq0KU1KOhJOS6FQJFlPVlc/e1LdWiRDWlpa5MuXGz+/p4SFhanl5cmTk/s+F1m6bD39NOzd/g5fn5/53U0QQmN6GaLfV62pYXnsEnzt9EdbYy/0C0mPUSQahULB1ctHuP6vW6R51UEDewFE+x5IIcSfTeYYhYjGihUbGTSoF/9eO86Rwyf59u0blSuXo2LFMhw5chIXl8ibvIUQf77kOCSaUBIYRaIaPnIy3nfv07VLW9q3b4menh4PHvoxYuRk5s5b8bubJ4RIIslx20VCSWAUiUqhULB23TbWrou8bF8IkXKlnLAoc4xCCCGEGukxCiGE0JjMMYoklTVrZm56nWL8hNksWLgqxrK9e3VkwfzJdO7iyIaNztGW09LS4vzZfTx/8YrmLbokdpPFX+jtu/csXr0Z9wuXeff+I2lNjKlYtiR9uzqQM3tWZbmdrocZN31+lPewLFKQLSvnKb++cOVfug2IejtP+nRmnN63BYBRk2ZHetPGfzWpX4fJToPi+VQiof57bu6fTAJjMmNomIadzqtIm9Yk1rK5cmVn8qQRcbrvvLkTKVeuFHtdD2vaRCF4++49bboN4OWrN1QqV4r6tavz6PFTDh47xdmLV9myYi65c0bsY/bxfQhAl3Yt0f/PmzoyZ8yg9vWPsi2bNCBDejO1PNVN9rWsKpEtmtdh7dx7iDfv3lP2P8cbiqQlPUaRJHLlys4O51XRHtj9X8uWzIj1jRWpUqVi2dIZtGsb+ag3IRJq8erNvHz1hiH9utGhTTNl+v4jJxg+YSYzF65k0YxxQESwS2tijGOvzrHe1+d+RGAc2LtzjC8Hrm1VmdpWkV8kfvTkGd68e08D6xo0bVg3nk8lNJGSVqXK4ptkon+/rly/5kYJyyKcOHE21vId2reibt0aHDrkFm2Z2rWq4XX9BO3aNufo0VOJ2Frxt3NzP08607Q4tLJVS29kU4uc2bNy7rIH4eERfYh7vo8wz58nTvf18X1ItiyZYgyK0fn46TPjZyzENK0JIx17xft6oZmUtMFfAmMy0b9fV/weP6VmreZs2uwSY9ksWTIxa+ZY1m9w5tjxyC8X/sHevhnGxkZ07TaQXn2GJXaTxV/q27dvdGvfmt5d2qGtHflXiL6eHqGhYYSGhvHy9Rs+ffbHIn/eON33waMncSoblWVrt/Dpsz99uzlgGoepCJG4UtJrp2QoNZno3WcYx93OEB4ejrl5zOcZLlo4hZCQUAYPGY9DuxbRlluzZgsDHEfj7x9A7twxvxtSiLjS0dGJ1FP84YHfEx4+fkrO7FkxMNBXDo2GhYXRf8QErnvdJig4hJLFC9OvW3uKFymovPbR46cEh4RgYGDA8AkzuezhyWf/AAoXzE+PDnZUrVg22jY9e/GK7XsOkCNbFlo0rp+ozyv+PtJjTCaOHjutHHqKScuW/2DbpD4DBo5RvhQ5OufOX8HfPyCRWihEzMLDw5kyZwnh4eG0bBIRnHx8HwHgvOcgwcEh2Da0plK5UlzyuE773oM5d8lDef3d7wtvjpxw59mLlzSsW5NaVpW4c9eXXoPHsGv/kWjr3rxjL6GhYTi0skVXVyfpHlJEKyW9XUN6jH+QdOnMmD93IvsPHGPHDtff3RwhlBQKBeNnLOTi1esULWSu7FGGh4eTLUsm+nfvoPZOwyv/etH1fyNwmjyHwzvWYmCgT3BwCDmzZ6V543p0dfj5Kirfh3607TGQKXOWYlW5PBnSqa9W/fI1iD0Hj5HWxJimjWx+yfOKyFLSdg3pMf5B5s2dQKpUBvTpG7ctGkL8CmFh3xg9ZS4u+w6TI1sWFk4bq3y7SvcObTjqsj7Si37LlbKkoXVN3rx7z9XrNwBo2rAuh5zXqAVFgPx5c9OulS1BwcGccL8Qqf6TZy7w2T+ABnVqyHsTf6OU1GOUwPiHaNigDvZ2zRg5airPnr343c0RAoCvQUH0Gz6ePQePkTtndtYunE6mjOnjdG3hggUAePr8Zaxli3wv++xF5LInz14EwLpm1bg2WyQBhQb/S25iHEp1cHAgNDRULU2hUKClpcW2bXJI9K/UrFlDIGLhzaKFUyLlr1k9lzWr51K7TgtOR/FXtRCJ7dNnf3oNGo3X7bsUtsjPsjmTSG9mqlbm9t37fPn6lbIlI2+2Dw4OAcDg+6Z/34d+vH77noplS6KlpaVWNig4GCDSAQHfvn3j3CUP0pmmpUyJoon1aCIBkmPPL6FiDIyDBw/GycmJxYsXo6MjE9q/017Xw/j5PYmUXqF8aWxsarLX9TCenrd4FEUZIRJbcHAIfYaOw+v2XcqWKs6i6WMxMoy897D/iAm8fvOO0/u2YGaaVi3vmtctAIoWNgdgwsxFeHjexHnNQmUP8Yd/PW9HlC1krpb+0O8p/gGB1LKqJL+jfrNwRfLr+SVUjIGxRIkSNGnShLt372Jtbf2r2iSi4Op6BFfXyKvy+vfrGhEY9x6J8axUIRLT/OXruH7jNiWKFWbZ7ImkMjCIspxNzWqs37aL+cvXMXZof2VP8MiJM7ifv0zZksUwz5cHgLo1q+LheZOFK9azcPo45erSf2/cZue+Q+TMnpWqFdS3bNy55wtAsUIWSfSk4m8U66rUrl27/op2CCH+EG/fvWfrrn0A5MuTk9WbdkRZrmu7VvToaMeZi1fY6XoYn/sPKVWiKI8eP8X9/BUypk/HxJEDleVbNW3IsVNnOXPxKi069qFKhTK8fP0GN/cL6OvpMWPcsEhbMZ58n2/PlSNbEj2tiKuU01+U7RpCiHjyvOVNaGgYALv3H422nEMrW0yMjdi0bA5L12zm+OnzbN7hillaE5o1sqFvVwcyZkinLK+nq8uKuZNZudGZg8dOsXmnK8ZGaahTvTJ9uzqQJ1fkQyo+fvoMQOZMGSLliV8rOZ5gk1BaCkXSDQzr6mdPqlsL8Ut9fX7mdzdBCI3pZYj5VC1N2OW2TfC1W/32JFo7EoP0GIUQQmjsr1mVKoQQQsRFShpKlcAohBBCY8lxo35Cyck3QgghhArpMQohhNCYzDEKIYQQKpJwg8MvJ0Opv8iMaaMJC3lGdatK8b62YYM6hIU8o0Q0Z0Fqa2vTp3cnrnkcw//TfR7cv8zKFbPJmjVzpLIjR/yP5089efXiJmtWzyNtFG86r2tdnZCgJ1haFol3W8WfqViV+rF+Ll/zUpbf6Xo42nL23QYkqA2OoyZRrEp9nr14FSlvwYr10dY3eMxUtbI7XQ9Ru6kDFes2p9+w8bx68zbS/XwfPcayWkNOfT+AXGguHEWCP8mN9Bh/gXJlS9K/f8JOECpUqACrVs6JsczaNfNoa98cj2teLFmyjoKFCtCpYxuqW1WiXIV6fPq+CdrWtj4Txg/l6NFT3Pd9RKeOrUmTJjVt7Hqo3W/8+KHsdNmPl9ftBLVZ/Hl6dW4bZfr7Dx/ZvvsA6cxMyZf75wZ7n+8vFe7SrmWkg70zZ4z/Zvtjp85x7NS5aPN97j9EX1+PLu1aRcozz5db+f9ved9j3PQFWBYpSIkaVdl76Dj/GzGRrSvnqR1MvmjlBooVMqdG1YrxbquImgylijjT09NjxYpZ6OrG/1tdo3pltm5ZRsYYXuPTrFlD2to3Z9v2PTi076sczhjo2IMZ08fQr29nJk2eB0DnjnbcvuNDg0YRvwSfPn3OlMkjSZfOjPfvPwDQpEk9SpUsRsdO/ePdXvHn6tOlXdTpQ8cCMHX0YDKk/3lKjY/vQ9KaGOPYq7PGdX/67M+k2YtjLOPj+5D8eXJF284fdh84iomxEWsWTSeVgQHFixZk6Njp3L57X3kA+e279zl++jzL50zSuO3iJ1mVKuJs5Ij+WJjn4/hx9zhfkypVKpYvm8mRw9vQ0tLCQ2UI67/69u7E58/+9Os/Sm2Mf9nyDWzctJPXr98p0/LkzcmNG3eUX3t6RrzdIG+enABoaWkxbuxgNm/Zxd27vnFur0iZ9hw4xulzl7FtYE2VCmXU8u75PsI8f55EqWf6/OWEhoZSomihKPMDAgN5/vI1FvnzxnqvZ89fkidnduWh5oXM80ekq7zHceGK9ZQpWYzK5UsnQuvFDylpKFUCYxIqXrwww4b2ZfqMRdy67RPn6zJnzkCXzvYcPORG6bLW3LzpHWU5Q8M0VKlSnlOnz/Phw0e1vC9fvtKp8/9YsXKjMu3jh09qrwYyNjEGIv5iB2jTxpaCFvmZOCnmoVuR8n0NCmLBivWkSZ0ax97qvcKXr9/w6bN/nAJVbM5evIrrYTeG9utO+nRmUZbxuR8xbGtRIPb6TIyN+PI1SPl1YOAXAOXP/TWvW5y5eJX+3dpr2nSRgklgTCLa2tqsXDGbe/cfMnXawnhd++HDJ6yqN6Fps048j+Ht5kUKW6Cjo8Pt2z5UKF+awwe38uHdXV4+v8GK5bPIoHJAM8ClS9eoVasKlSqWJX16M3r1aM/z5y95+PAxOjo6jHEaxNp123n0SN7p+LfbuH0Pr9++o31r20gvH/4RqMLCwug/YgJWDdtQvk4zujuO4sbtu3GuIzDwC+NmLKBi2VLYNoz+tXY+vo8A+PDxM13/N5LK9VpSuV5LHEdN4qHfU7WylkULcf+hH8dPnyMgMJAN23eTOpUBhcwjzghdsGI9VSqUoXSJYnFup4gbhUKR4E9yI4ExiQwa2JNSJYvRo8cQQkND43Xt58/+nL9wNdZyWbNFrDotVbIYp07uwsBAn9Wrt3Dnjg+dO9lx+tQeTL73CgFmzFrM8+evOOO+l1cvblKhQmn6/W8U3759o2OH1uTMmZUpU+cry//3Leri7xAaGsoWF1cM9PWxb/FPpPwfgcp5z0GCg0OwbWhNpXKluORxnfa9B3Pukkec6pm9ZDWfPn1m7NB+MZa7+z0Qr9uyEyPDNDRvXI/iRQpy7NQ57LsPwNvn57B/i3/qU6SgOQNGTqJi3RYcOXEGx16dSWdmyvnL17j67w36dnNQlg8PT0lLRn6vcA0+yY0svkkC5ub5GDN6IEuXrediHH9JJIShYRoAbGxqMnXaAkaPma7Mmzl9DI6OPRg7ehCDhowD4M2bd5Qtb4OtbX1MjI04dtwdHx9f9PX1GTVyAMuXb+TZsxfUr1eLBfMnkzt3DjyuedG9x2C1uUmRsh0+cYa37z7Qskl90v2ntwgRwSRblkz0796BRja1lOlX/vWi6/9G4DR5Dod3rMXAQD/Stapld+w9xKA+XciZPWuM7dHR0SZblkxMGjWI8qUtlen7j5xg+ISZjJ46lx1rFwFgYKDP5uWzOX76PG/evqdsqeIUKVgAgIUrN1DLqhLFCxfkxu27jJ4yF99Hj8mTKwejB/dVu7eIP1l8I2K0cvksXr9+xyinqbEX1sCPv3ZfvXrD+Amz1fLGjp9JQEAgLVo0Ukv39w9g48YdLF6yFp/vf2n36O5AunSmTJuxkPTpzdi2dTlXrl6nUeN2fAn8wo7tKxO0qlb8mVwPHQegeeN6UeZ379CGoy7r1YIiQLlSljS0rsmbd++5ev1GtPcPCg5m7LT5FClYAIdWtrG2x2lQH466rI8UuBrZ1KJsyWLc8fFVG1LV09Ojfp3qtG/TVBkUT5y5wM07PvTt6kBoWBgDRk0inZkpy2ZPxCJ/HgaMnKicaxcJI4tvRLR69+pI1aoV6NtvhHLiP6l8/hTxH/LNm96EhYWp5X358pV79x+SPXtWDL6v0ItKmjSpGTa0L4uXrOHNm3fYtWmKgYE+vfsM5+ix0wwYOIYCBfJiU7dGUj6KSCYCAgO58u8NsmfNTLHCFvG+vvD3QPQ0hrnxhSs38PzFKyaMcERHRyfBbQUobBFRn+qq0/9SKBQsXLmBerWtsMifl7MXrvLq9VuGD+hBlQplGDOkH4FfvnDg6EmN2vK3S0lzjNINSGTNmzUEYJ/rxijz3Y7vBCC/eQX8/rNwIL7ufZ970dfXizJfT0+X0NBQQkJCor1Hv75dSJ06FbNmLwOgQIG8vH37no8fPwHg4/MAgHwqm6hFynXh8r+EhYVRp3qVaMvcvnufL1+/UrZk8Uh5wcERP2sG+tEPox47eZawb99o3qF3lPk2LToCcPPcIcLCvnHn3n0U4Qoso9jOEfS9vv8eMqDq0PHTPHj0mLmTRgHg9/QZALlzRLxIPa2JMaZp0/Lk2Yto7yFilxx7fgklgTGRrd+wg9PuFyKl29StSYUKpVm/wRk/vyd8/PhZ47p8fR/x/PlLypQpgZGRIQEBgcq8tGlNyJ8vN7du+0T7F5mJiTGDBvZk7rzlyu0eurq6asOmqVJF9DaT4191IvF53orYGlSmZPSrNvuPmMDrN+84vW8LZqZp1fKueUXsjS1a2Dza6x1a2fJZ5Wf1h8PHT/Pw8VPatWyCsbERAOHh33DoOZg0qVNx5sA2tR6mQqHg+s3b6OroKFed/ldY2DcWr95E43q1yZMr4uSeb9++qf0TICQkRBabCSUJjIlsw0bnKNPTpk1LhQql2bDBOcrAmRAKhYI1a7fiNMqRmTPG0Kv3MGXepInDSZ06NevWbYv2+kEDexIermDe/JXKNB8fXzJkSEeBAnm5f/8hFStEbIK+d+9BorRZJG/e9yLmnYsVin4Y1aZmNdZv28X85esYO7S/MqAcOXEG9/OXKVuyGOb58kR7vUPrplGm3733gIePn+LQuinZv5/zq6+vT40qFTh++hyrNjnTo4Odsvy6rS7c833EP/VqY/I9kP7X3kPHefbiFSvmTlam/QiQnre8qVSuFA/8nuAfEEjunNmjbbOIXVIuvrG1tcXYOGKFfY4cOejZsyfDhw9HS0sLc3Nzxo4di7a2Ns7Ozmzbtg1dXV169epFzZo1E1SfBMbfbMzogQBMmJiwTfXTpi+iTm0runVth2XxIpw9e4kKFUpTtWoFzp69xNJl66O8LkOGdPTv15XJU+ap9TSdd7gyftwQ9u3dyMGDx2nTxpY73vc4Fo+Te8Sf68mzF6QyMCBTDMcQ9uhox5mLV9jpehif+w8pVaIojx4/xf38FTKmT8fEkQPVym/cvpvPAYE4tLKNNoDFZEi/bly/eZuFKzZw5doNCprn5bb3fa7860W+PDkZ2r97lNeFhoaybO0Wmje2UQZagCoVypA5UwZGTJxJgzo1OHn2IqZpTWgo8+gaCU+iUaXg4GAANm78OT3Vs2dPBgwYQIUKFRgzZgxubm6ULFmSjRs34uLiQnBwMPb29lSpUiXGYfboyOKb32zM6EGMGT0owdcHBQVhbdOKSZPnkj69GX37diZ79qxMm76Qeg3s1YaLVA0b0peAgEAWL1mrlv7y5Wua2HYgOCSYnj3b4+PjS/MWXWS/11/i46fPZM4U8yHgJsZGbFo2B4dWtrx594HNO1y57X2fZo1scF6zMNL2i43Oe1i6ZjP+UQyfxkX2rJnZvnoBTRvV5f7DR2ze4cqzFy/pYNeMzcvnYhrFG2IgYp/l+w8f1XqZAKkMDFg6awLZs2Rm6659GKZJzaIZ49ROhRLxp9DgExNvb2++fv1K586dad++PdevX+fWrVuUL18eACsrK86fP4+XlxelSpVCX18fY2NjcuXKhbd31KeGxUZLkYSTR7r6MjQhUoavz8/87iYIoTG9DFHPxSaGKtlrxV4oGueenYg27+7du3h6etKyZUsePXpEt27dCAoK4uzZswBcuHABFxcXqlWrho+PD0OGDAFg6NCh2NraUrly5Xi3R4ZShRBCaCypVqXmzZuX3Llzo6WlRd68eTE1NeXWrVvK/MDAQExMTDAyMiIwMFAt/ce8ZHzJUKoQQgiNJdU+xp07dzJt2jQAXr16RUBAAFWqVOHSpUsAuLu7U7ZsWSwtLfHw8CA4OBh/f398fX2xsIj/XlyQHqMQQohkrEWLFowYMQI7Ozu0tLSYMmUKZmZmjB49mjlz5pAvXz5sbGzQ0dHBwcEBe3t7FAoFjo6OMR5uEhOZYxQiDmSOUaQESTnHWD5b9QRfe/n56URsieakxyiEEEJjKekQcQmMQgghNJaSTseSwCiEEEJjclaqEEIIoUJ6jEIIIYSKlNRjlH2MQgghhArpMQohhNCYrEoVQgghVCTV2zV+BwmMQgghNCY9RiGEEEKF9BiFEEIIFdJjFEIIIVSkpB6jbNcQQgghVEiPUQghhMZkKFUIIYRQkZKGUiUwCiGE0Jj0GIUQQggVCkX4725CopHAKIQQQmMp6RBxCYxCCCE0lpJeOyXbNYQQQggV0mMUQgihMRlKFUIIIVSkpKFUCYxCCCE0JvsYhRBCCBWyj1EIIYRQIUOpQgghhIqUtPhGtmsIIYQQKqTHKIQQQmMylCqEEEKokFWpQgghhArpMQohhBAqUtLiGwmMQgghNCY9RiGEEEJFSppjlO0aQgghhArpMQohhNCYHAknhBBCqEhJQ6kSGIUQQmhMFt8IIYQQKmQoVQghhFAhPUYhhBBCRUoKjLJdQwghhFAhPUYhhBAaSzn9RdBSpKT+rxBCCKEhGUoVQgghVEhgFEIIIVRIYBRCCCFUSGAUQgghVEhgFEIIIVRIYBRCCCFUSGD8Q4WHhzNmzBhat26Ng4MDfn5+v7tJQiSYp6cnDg4Ov7sZQgCywf+Pdfz4cUJCQti+fTvXr19n2rRpLF269Hc3S4h4W7lyJa6urqROnfp3N0UIQHqMfywPDw+qVasGQMmSJbl58+ZvbpEQCZMrVy4WLlz4u5shhJIExj9UQEAARkZGyq91dHQICwv7jS0SImFsbGzQ1ZXBK5F8SGD8QxkZGREYGKj8Ojw8XH65CCFEIpDA+IcqXbo07u7uAFy/fh0LC4vf3CIhhEgZpIvxh7K2tubcuXO0adMGhULBlClTfneThBAiRZC3awghhBAqZChVCCGEUCGBUQghhFAhgVEIIYRQIYFRCCGEUCGBUQghhFAhgVEIIYRQIYFRCCGEUCGBUQghhFDxfwKj1gxvz0KHAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictions = lr.predict_proba(X_test)[:,1] >= 0.664\n",
    "\n",
    "print(\"The recall score is: %.2f\" % recall_score( y_test, predictions))\n",
    "print(\"The precision score is: %.2f\" % precision_score( y_test, predictions),\"\\n\")\n",
    "print(\"Accuracy score is: %.2f\" % accuracy_score( y_test, predictions))\n",
    "print(\"The F1 score is: %.2f\" % f1_score( y_test, predictions))\n",
    "print(\"The balanced accuracy is: %.2f\" % balanced_accuracy_score(y_test, predictions))\n",
    "\n",
    "groupNames = [\"True Neg\", \"False Pos\", \"False Neg\", \"True Pos\"]\n",
    "groupCount = [\"{0:0.0f}\".format(value) for value in\n",
    "                cm.flatten()]\n",
    "groupPercent = [\"{0:.2%}\".format(value) for value in\n",
    "                     cm.flatten()/np.sum(cm)]\n",
    "labels = [f\"{v1}\\n{v2}\\n{v3}\" for v1, v2, v3 in zip(groupNames,groupCount,groupPercent)]\n",
    "labels = np.asarray(labels).reshape(2,2)\n",
    "\n",
    "cm = confusion_matrix(y_test , predictions )\n",
    "sns.heatmap(cm, annot=labels, annot_kws={\"size\": 20}, fmt=\"\", cmap=\"rocket\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48000a44",
   "metadata": {},
   "source": [
    "# Summary <a id=\"summary\"></a>\n",
    "\n",
    "Tracking customer sentiment can help businesses improve products and services. In this project, I was able to build models to deal with two separate business scenarios in which predicting sentiment could be useful to airline companies.  For tracking customer sentiment, I was able to get an accuracy of 92% and a balanced accuracy score of 88% by optimizing the balanced accuracy score to determine the best threshold for the model. For identifying customers with negative experiences, I was able to get an F1-score of  95%. The next steps would include looking deeper at the causes of misclassification to improve the model. One thing I noticed is that a tweet can have positive language, but the user can be using sarcasm which can throw the model off. With more time, I would look deeper into this. An issue with the dataset is that all the  tweets collected are from Feb 23, 2015 and Feb 22, 2015. It’s likely that collecting more tweets would yield a better and more generalizable model. I would be interested in what a whole year of customer sentiment would look like. In particular this might help with anomalous periods such as holidays and summers where there may be more customers and higher negative sentiment. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ae0eb112",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}