predicting-sentiments-with-logistic-regression.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"source":"<a href=\"https://www.kaggle.com/code/arunachal/predicting-sentiments-with-logistic-regression?scriptVersionId=139734234\" target=\"_blank\"><img align=\"left\" alt=\"Kaggle\" title=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"></a>","metadata":{},"cell_type":"markdown"},{"cell_type":"markdown","source":"*The dataset used here contains text reviews of a popular web-service, alongwith pre-coded sentiment types. We explore the distributions of scores, sentiment types and review lengths. And compare performance of 4 models (Count Vectorizer-Binary, Count Vectorizer-Non-Binary, CountVectorizer-NGrams, TfidfVectorizer). Finally we use the best model to predict sentiment types for 2 test cases: a positive and a negative text review.*","metadata":{}},{"cell_type":"code","source":"import os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-08-12T19:26:47.805944Z","iopub.execute_input":"2023-08-12T19:26:47.806265Z","iopub.status.idle":"2023-08-12T19:26:47.819398Z","shell.execute_reply.started":"2023-08-12T19:26:47.80623Z","shell.execute_reply":"2023-08-12T19:26:47.818756Z"},"trusted":true},"execution_count":1,"outputs":[{"name":"stdout","text":"/kaggle/input/product-reviews/Canva_reviews.xlsx\n","output_type":"stream"}]},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport seaborn as sns\nimport matplotlib.pyplot as plt\n%matplotlib inline\n\nfrom collections import Counter\nimport pickle\n\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.metrics import accuracy_score\n\nimport nltk\nfrom nltk.corpus import stopwords\nfrom nltk.stem import PorterStemmer, LancasterStemmer\nfrom nltk.tokenize import sent_tokenize, word_tokenize, RegexpTokenizer\n\nnltk.download('stopwords')","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:47.820545Z","iopub.execute_input":"2023-08-12T19:26:47.821005Z","iopub.status.idle":"2023-08-12T19:26:48.804375Z","shell.execute_reply.started":"2023-08-12T19:26:47.820977Z","shell.execute_reply":"2023-08-12T19:26:48.803662Z"},"trusted":true},"execution_count":2,"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.5\n  warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n","output_type":"stream"},{"name":"stdout","text":"[nltk_data] Downloading package stopwords to /usr/share/nltk_data...\n[nltk_data]   Package stopwords is already up-to-date!\n","output_type":"stream"},{"execution_count":2,"output_type":"execute_result","data":{"text/plain":"True"},"metadata":{}}]},{"cell_type":"code","source":"df = pd.read_excel('../input/product-reviews/Canva_reviews.xlsx')\ndf.head()","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:48.805394Z","iopub.execute_input":"2023-08-12T19:26:48.805891Z","iopub.status.idle":"2023-08-12T19:26:49.644588Z","shell.execute_reply.started":"2023-08-12T19:26:48.80586Z","shell.execute_reply":"2023-08-12T19:26:49.643762Z"},"trusted":true},"execution_count":3,"outputs":[{"execution_count":3,"output_type":"execute_result","data":{"text/plain":"                                            reviewId            userName  \\\n0  gp:AOqpTOFxf3fttcT5DSvFIn9KPp5FErgH9yC533Fmoxv...      Donna Caritero   \n1  gp:AOqpTOEq6rNIWLnPV4KFTctWvm0mpGEQljtD6mvy1H-...  Soumi Mukhopadhyay   \n2  gp:AOqpTOE86hSyPRHZgYt28Uk5zGe4FZGb1hkmtFDiYJ2...   Theknown _unknown   \n3  gp:AOqpTOHSuKkVTcM3QgCCKysHQlxEnk2ocOKsUMiMIJy...        Anthony Dean   \n4  gp:AOqpTOEOrZt5H6jXPiplJyffCd5ZBnVXACTWgwNsF1R...   Neha Diana Wesley   \n\n                                           userImage  \\\n0  https://play-lh.googleusercontent.com/a-/AOh14...   \n1  https://play-lh.googleusercontent.com/a-/AOh14...   \n2  https://play-lh.googleusercontent.com/a-/AOh14...   \n3  https://play-lh.googleusercontent.com/a/AATXAJ...   \n4  https://play-lh.googleusercontent.com/a-/AOh14...   \n\n                                              review  score  thumbsUpCount  \\\n0  Overall it's really an amazing app. I've been ...      4            528   \n1  Hey! Yes I gave a 5 star rating... coz I belie...      5            351   \n2  Canva used to be a good app! But recently I've...      1            160   \n3  It's a brilliant app, but I have just one prob...      5            145   \n4  This was such a great app. I used to make BTS ...      4            142   \n\n  reviewCreatedVersion                  at  \\\n0              2.116.0 2021-06-17 07:18:54   \n1              2.116.0 2021-06-17 19:18:28   \n2              2.116.0 2021-06-23 19:13:28   \n3              2.116.0 2021-06-19 23:36:07   \n4              2.116.0 2021-06-21 00:29:27   \n\n                                        replyContent           repliedAt  \\\n0  Hi Donna. We are sorry that your text or desig... 2021-06-19 21:24:32   \n1                                                NaN                 NaT   \n2  Hi there. We're sorry to hear that you are hav... 2021-06-26 20:20:56   \n3                                                NaN                 NaT   \n4  Hi Neha. Sorry to hear that you are unable to ... 2021-06-24 20:46:28   \n\n  Sentiment              Sub Category    Sub Category_test  \n0  Negative                       NaN  bug_picture_quality  \n1  Positive           extremely_happy                  NaN  \n2  Negative  saving_downloading_issue     bug_saving_files  \n3  Negative                    others        bug_app_crash  \n4  Negative                    others            bug_other  ","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>reviewId</th>\n      <th>userName</th>\n      <th>userImage</th>\n      <th>review</th>\n      <th>score</th>\n      <th>thumbsUpCount</th>\n      <th>reviewCreatedVersion</th>\n      <th>at</th>\n      <th>replyContent</th>\n      <th>repliedAt</th>\n      <th>Sentiment</th>\n      <th>Sub Category</th>\n      <th>Sub Category_test</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>gp:AOqpTOFxf3fttcT5DSvFIn9KPp5FErgH9yC533Fmoxv...</td>\n      <td>Donna Caritero</td>\n      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n      <td>Overall it's really an amazing app. I've been ...</td>\n      <td>4</td>\n      <td>528</td>\n      <td>2.116.0</td>\n      <td>2021-06-17 07:18:54</td>\n      <td>Hi Donna. We are sorry that your text or desig...</td>\n      <td>2021-06-19 21:24:32</td>\n      <td>Negative</td>\n      <td>NaN</td>\n      <td>bug_picture_quality</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>gp:AOqpTOEq6rNIWLnPV4KFTctWvm0mpGEQljtD6mvy1H-...</td>\n      <td>Soumi Mukhopadhyay</td>\n      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n      <td>Hey! Yes I gave a 5 star rating... coz I belie...</td>\n      <td>5</td>\n      <td>351</td>\n      <td>2.116.0</td>\n      <td>2021-06-17 19:18:28</td>\n      <td>NaN</td>\n      <td>NaT</td>\n      <td>Positive</td>\n      <td>extremely_happy</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>gp:AOqpTOE86hSyPRHZgYt28Uk5zGe4FZGb1hkmtFDiYJ2...</td>\n      <td>Theknown _unknown</td>\n      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n      <td>Canva used to be a good app! But recently I've...</td>\n      <td>1</td>\n      <td>160</td>\n      <td>2.116.0</td>\n      <td>2021-06-23 19:13:28</td>\n      <td>Hi there. We're sorry to hear that you are hav...</td>\n      <td>2021-06-26 20:20:56</td>\n      <td>Negative</td>\n      <td>saving_downloading_issue</td>\n      <td>bug_saving_files</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>gp:AOqpTOHSuKkVTcM3QgCCKysHQlxEnk2ocOKsUMiMIJy...</td>\n      <td>Anthony Dean</td>\n      <td>https://play-lh.googleusercontent.com/a/AATXAJ...</td>\n      <td>It's a brilliant app, but I have just one prob...</td>\n      <td>5</td>\n      <td>145</td>\n      <td>2.116.0</td>\n      <td>2021-06-19 23:36:07</td>\n      <td>NaN</td>\n      <td>NaT</td>\n      <td>Negative</td>\n      <td>others</td>\n      <td>bug_app_crash</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>gp:AOqpTOEOrZt5H6jXPiplJyffCd5ZBnVXACTWgwNsF1R...</td>\n      <td>Neha Diana Wesley</td>\n      <td>https://play-lh.googleusercontent.com/a-/AOh14...</td>\n      <td>This was such a great app. I used to make BTS ...</td>\n      <td>4</td>\n      <td>142</td>\n      <td>2.116.0</td>\n      <td>2021-06-21 00:29:27</td>\n      <td>Hi Neha. Sorry to hear that you are unable to ...</td>\n      <td>2021-06-24 20:46:28</td>\n      <td>Negative</td>\n      <td>others</td>\n      <td>bug_other</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"### Exploring Sentiment Counts, Score Counts, and Review Lengths","metadata":{}},{"cell_type":"code","source":"plt.figure(figsize=(20, 7))\n\nplt.subplot(131)\nsns.countplot(x='Sentiment', data=df)\nplt.title('Sentiment Frequencies')\n\nplt.subplot(132)\nsns.countplot(x='score', data=df)\nplt.title('Score Frequencies')\n\nplt.subplot(133)\nsns.countplot(x='score', data=df, hue='Sentiment')\nplt.title('Score and Sentiment Frequencies')","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:49.648471Z","iopub.execute_input":"2023-08-12T19:26:49.64898Z","iopub.status.idle":"2023-08-12T19:26:50.581121Z","shell.execute_reply.started":"2023-08-12T19:26:49.648951Z","shell.execute_reply":"2023-08-12T19:26:50.579952Z"},"trusted":true},"execution_count":4,"outputs":[{"execution_count":4,"output_type":"execute_result","data":{"text/plain":"Text(0.5, 1.0, 'Score and Sentiment Frequencies')"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 2000x700 with 3 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"df['length'] = df['review'].apply(len)\ndf['length'].describe()","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:50.582971Z","iopub.execute_input":"2023-08-12T19:26:50.583304Z","iopub.status.idle":"2023-08-12T19:26:50.596943Z","shell.execute_reply.started":"2023-08-12T19:26:50.583276Z","shell.execute_reply":"2023-08-12T19:26:50.595801Z"},"trusted":true},"execution_count":5,"outputs":[{"execution_count":5,"output_type":"execute_result","data":{"text/plain":"count    1500.000000\nmean       86.468667\nstd        55.681355\nmin        20.000000\n25%        54.000000\n50%        75.000000\n75%       102.000000\nmax       498.000000\nName: length, dtype: float64"},"metadata":{}}]},{"cell_type":"code","source":"sns.histplot(df['length'])","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:50.597977Z","iopub.execute_input":"2023-08-12T19:26:50.598828Z","iopub.status.idle":"2023-08-12T19:26:51.009644Z","shell.execute_reply.started":"2023-08-12T19:26:50.598796Z","shell.execute_reply":"2023-08-12T19:26:51.008501Z"},"trusted":true},"execution_count":6,"outputs":[{"execution_count":6,"output_type":"execute_result","data":{"text/plain":"<Axes: xlabel='length', ylabel='Count'>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"sns.displot(x='length', data=df, hue='Sentiment', kind='kde', fill=True)\nplt.title('Distribution of Review Length for Different Sentiment Types')","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:51.011163Z","iopub.execute_input":"2023-08-12T19:26:51.011543Z","iopub.status.idle":"2023-08-12T19:26:51.667905Z","shell.execute_reply.started":"2023-08-12T19:26:51.011511Z","shell.execute_reply":"2023-08-12T19:26:51.66673Z"},"trusted":true},"execution_count":7,"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n  self._figure.tight_layout(*args, **kwargs)\n","output_type":"stream"},{"execution_count":7,"output_type":"execute_result","data":{"text/plain":"Text(0.5, 1.0, 'Distribution of Review Length for Different Sentiment Types')"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 612.5x500 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"sns.displot(x='length', data=df, hue='score', kind='kde', fill=True)\nplt.title('Distribution of Review Length for Different Score Types')","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:51.669511Z","iopub.execute_input":"2023-08-12T19:26:51.670677Z","iopub.status.idle":"2023-08-12T19:26:52.489769Z","shell.execute_reply.started":"2023-08-12T19:26:51.670643Z","shell.execute_reply":"2023-08-12T19:26:52.488584Z"},"trusted":true},"execution_count":8,"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n  self._figure.tight_layout(*args, **kwargs)\n","output_type":"stream"},{"execution_count":8,"output_type":"execute_result","data":{"text/plain":"Text(0.5, 1.0, 'Distribution of Review Length for Different Score Types')"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 558.875x500 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"### Text Cleaning","metadata":{}},{"cell_type":"code","source":"# Tokenizing\n\nreviews = list(df['review'])\nreviews_lower = [r.lower() for r in reviews]\ntokens = [word_tokenize(r) for r in reviews_lower]\n\n# Removing stop words\n\nsw = stopwords.words('english')\ntokens = [[word for word in t if word not in sw] for t in tokens]\n\n# Removing special characters\n\nregex_tokenizer = RegexpTokenizer(r'\\w+')\n\ntokens = [[\"\".join(regex_tokenizer.tokenize(word)) for word in t \n           if len(regex_tokenizer.tokenize(word)) > 0] for t in tokens]\n\n# Stemming\n\nporter = PorterStemmer()\nlancaster = LancasterStemmer()\n\n# porter.stem('teachers'), lancaster.stem('teachers'), porter.stem('absolutely'), lancaster.stem('absolutely') # porter performs better\n\ntokens = [[porter.stem(word) for word in t] for t in tokens]\nflat_tokens = [word for t in tokens for word in t] # flattening the token list\n\ncounts = Counter(flat_tokens)\nprint('Most frequent tokens: \\n\\n', counts.most_common(10))\n\nclean_reviews = [' '.join(t) for t in tokens]\nprint('\\n\\nClean Review Count: ', len(clean_reviews))\n","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:52.491148Z","iopub.execute_input":"2023-08-12T19:26:52.491491Z","iopub.status.idle":"2023-08-12T19:26:53.611889Z","shell.execute_reply.started":"2023-08-12T19:26:52.491461Z","shell.execute_reply":"2023-08-12T19:26:53.610619Z"},"trusted":true},"execution_count":9,"outputs":[{"name":"stdout","text":"Most frequent tokens: \n\n [('app', 928), ('use', 473), ('design', 349), ('easi', 293), ('s', 240), ('love', 220), ('good', 219), ('edit', 190), ('make', 190), ('great', 184)]\n\n\nClean Review Count:  1500\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Building, Testing, Saving Model 1: Binary Count Vectorizer","metadata":{}},{"cell_type":"code","source":"# Vectorizing clean reviews with binary CountVectorizer\n\nvect = CountVectorizer(binary=True, min_df=5)\nX = vect.fit_transform(clean_reviews)\nX.shape, len(vect.vocabulary_)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.613182Z","iopub.execute_input":"2023-08-12T19:26:53.613521Z","iopub.status.idle":"2023-08-12T19:26:53.649277Z","shell.execute_reply.started":"2023-08-12T19:26:53.613491Z","shell.execute_reply":"2023-08-12T19:26:53.648087Z"},"trusted":true},"execution_count":10,"outputs":[{"execution_count":10,"output_type":"execute_result","data":{"text/plain":"((1500, 414), 414)"},"metadata":{}}]},{"cell_type":"code","source":"# Visualising vector in array form and validating a test case\n\nx_array = X.toarray()\nx_array[18, :].sum(), clean_reviews[18], print(x_array)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.650837Z","iopub.execute_input":"2023-08-12T19:26:53.651287Z","iopub.status.idle":"2023-08-12T19:26:53.661908Z","shell.execute_reply.started":"2023-08-12T19:26:53.651256Z","shell.execute_reply":"2023-08-12T19:26:53.660792Z"},"trusted":true},"execution_count":11,"outputs":[{"name":"stdout","text":"[[0 0 0 ... 0 0 1]\n [0 0 0 ... 0 0 0]\n [0 0 0 ... 0 0 0]\n ...\n [0 0 0 ... 0 0 0]\n [0 0 0 ... 0 0 0]\n [0 0 0 ... 0 0 0]]\n","output_type":"stream"},{"execution_count":11,"output_type":"execute_result","data":{"text/plain":"(17,\n 'nt open design even let make new one sent video screenshot canvameandroid pleas look asap ve lot project social media site client project freelanc site pend',\n None)"},"metadata":{}}]},{"cell_type":"code","source":"# Converting target variable to numerical type\n\ndf['Sentiment'] = df['Sentiment'].apply(lambda x: 1 if x == 'Positive' else 0)\ny = df['Sentiment']","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.663512Z","iopub.execute_input":"2023-08-12T19:26:53.663828Z","iopub.status.idle":"2023-08-12T19:26:53.674807Z","shell.execute_reply.started":"2023-08-12T19:26:53.663794Z","shell.execute_reply":"2023-08-12T19:26:53.673838Z"},"trusted":true},"execution_count":12,"outputs":[]},{"cell_type":"code","source":"# Splitting vector into train-test sets, validating shapes\n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42)\nX_train.shape, X_test.shape, y_train.shape, y_test.shape","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.679985Z","iopub.execute_input":"2023-08-12T19:26:53.680876Z","iopub.status.idle":"2023-08-12T19:26:53.690901Z","shell.execute_reply.started":"2023-08-12T19:26:53.680834Z","shell.execute_reply":"2023-08-12T19:26:53.690022Z"},"trusted":true},"execution_count":13,"outputs":[{"execution_count":13,"output_type":"execute_result","data":{"text/plain":"((1200, 414), (300, 414), (1200,), (300,))"},"metadata":{}}]},{"cell_type":"code","source":"model = LogisticRegression()                                                   # Instancing model\nmodel.fit(X_train, y_train)                                                    # Fitting Model 1\ny_train_pred = model.predict(X_train)                                          # Predicting with the training set\ny_test_pred = model.predict(X_test)                                            # Predicting with the test set\naccuracy_score(y_train, y_train_pred), accuracy_score(y_test, y_test_pred)     # Getting accuracy scores for train and test sets","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.692026Z","iopub.execute_input":"2023-08-12T19:26:53.6923Z","iopub.status.idle":"2023-08-12T19:26:53.727827Z","shell.execute_reply.started":"2023-08-12T19:26:53.692276Z","shell.execute_reply":"2023-08-12T19:26:53.727073Z"},"trusted":true},"execution_count":14,"outputs":[{"execution_count":14,"output_type":"execute_result","data":{"text/plain":"(0.9616666666666667, 0.8833333333333333)"},"metadata":{}}]},{"cell_type":"code","source":"# Saving the vector and the model (Model 1: Binary Count Vectorizer)\n\nwith open(\"/kaggle/working/binary_count_vect.pkl\", \"wb\") as f:\n    pickle.dump(vect, f)\n    \nwith open(\"/kaggle/working/binary_count_vect_lr.pkl\", \"wb\") as f:\n    pickle.dump(model, f)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.728981Z","iopub.execute_input":"2023-08-12T19:26:53.729254Z","iopub.status.idle":"2023-08-12T19:26:53.736624Z","shell.execute_reply.started":"2023-08-12T19:26:53.729229Z","shell.execute_reply":"2023-08-12T19:26:53.73562Z"},"trusted":true},"execution_count":15,"outputs":[]},{"cell_type":"markdown","source":"### Building, Testing, Saving Model 2: Non-binary Count Vectorizer","metadata":{}},{"cell_type":"code","source":"# Casting text reviews as vectors with non-binary CountVectorizer and splitting vector into train and test sets\n\nvect = CountVectorizer(min_df=5)\nX = vect.fit_transform(clean_reviews)\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42)\n\n# Fitting Model 2 and getting prediction accuracy scores for train and test sets\n\nmodel = LogisticRegression()\nmodel.fit(X_train, y_train)\ny_train_pred = model.predict(X_train)\ny_test_pred = model.predict(X_test)\naccuracy_score(y_train, y_train_pred), accuracy_score(y_test, y_test_pred)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.737943Z","iopub.execute_input":"2023-08-12T19:26:53.738411Z","iopub.status.idle":"2023-08-12T19:26:53.802396Z","shell.execute_reply.started":"2023-08-12T19:26:53.738376Z","shell.execute_reply":"2023-08-12T19:26:53.801324Z"},"trusted":true},"execution_count":16,"outputs":[{"execution_count":16,"output_type":"execute_result","data":{"text/plain":"(0.9566666666666667, 0.8833333333333333)"},"metadata":{}}]},{"cell_type":"markdown","source":"Model 2 (non-binary) performs better than Model 1 (binary)!","metadata":{}},{"cell_type":"code","source":"# Saving the vector and the model (Model 2: Non-binary Count Vectorizer)\n\nwith open(\"/kaggle/working/count_vect.pkl\", \"wb\") as f:\n    pickle.dump(vect, f)\n    \nwith open(\"/kaggle/working/count_vect_lr.pkl\", \"wb\") as f:\n    pickle.dump(model, f)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.803565Z","iopub.execute_input":"2023-08-12T19:26:53.803906Z","iopub.status.idle":"2023-08-12T19:26:53.811064Z","shell.execute_reply.started":"2023-08-12T19:26:53.803871Z","shell.execute_reply":"2023-08-12T19:26:53.810018Z"},"trusted":true},"execution_count":17,"outputs":[]},{"cell_type":"markdown","source":"### Building, Testing, Saving Model 3: CountVectorizer with n-gram range 1-3","metadata":{}},{"cell_type":"code","source":"# Casting text reviews as vectors with CountVectorizer with 1-3 ngrams, and splitting vector into train and test sets\n\nvect = CountVectorizer(min_df=5, ngram_range=(1,3))\nX = vect.fit_transform(clean_reviews)\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42)\n\n# Fitting Model 3 and getting prediction accuracy scores for train and test sets\n\nmodel = LogisticRegression()\nmodel.fit(X_train, y_train)\ny_train_pred = model.predict(X_train)\ny_test_pred = model.predict(X_test)\naccuracy_score(y_train, y_train_pred), accuracy_score(y_test, y_test_pred)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.812926Z","iopub.execute_input":"2023-08-12T19:26:53.813265Z","iopub.status.idle":"2023-08-12T19:26:53.927774Z","shell.execute_reply.started":"2023-08-12T19:26:53.81323Z","shell.execute_reply":"2023-08-12T19:26:53.926666Z"},"trusted":true},"execution_count":18,"outputs":[{"execution_count":18,"output_type":"execute_result","data":{"text/plain":"(0.97, 0.8933333333333333)"},"metadata":{}}]},{"cell_type":"markdown","source":"Model 3 (n-grams) performs even better than Models 1 (binary) and 2 (non-binary)!","metadata":{}},{"cell_type":"code","source":"# Saving the vector and the model (Model 3: Count Vectorizer with 1-3 ngrams)\n\nwith open(\"/kaggle/working/ngram.pkl\", \"wb\") as f:\n    pickle.dump(vect, f)\n    \nwith open(\"/kaggle/working/ngram_lr.pkl\", \"wb\") as f:\n    pickle.dump(model, f)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.929417Z","iopub.execute_input":"2023-08-12T19:26:53.929863Z","iopub.status.idle":"2023-08-12T19:26:53.942784Z","shell.execute_reply.started":"2023-08-12T19:26:53.929823Z","shell.execute_reply":"2023-08-12T19:26:53.941614Z"},"trusted":true},"execution_count":19,"outputs":[]},{"cell_type":"markdown","source":"### Building, Testing, Saving Model 4: Tfidf Vectorizer","metadata":{}},{"cell_type":"code","source":"# Casting text reviews as vectors with TfidfVectorizer, and splitting vector into train and test sets\n\nvect = TfidfVectorizer(min_df=5)\nX = vect.fit_transform(clean_reviews)\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42)\n\n# Fitting Model 4 and getting prediction accuracy scores for train and test sets\n\nmodel = LogisticRegression()\nmodel.fit(X_train, y_train)\ny_train_pred = model.predict(X_train)\ny_test_pred = model.predict(X_test)\naccuracy_score(y_train, y_train_pred), accuracy_score(y_test, y_test_pred)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:53.944137Z","iopub.execute_input":"2023-08-12T19:26:53.94458Z","iopub.status.idle":"2023-08-12T19:26:54.008732Z","shell.execute_reply.started":"2023-08-12T19:26:53.944539Z","shell.execute_reply":"2023-08-12T19:26:54.007503Z"},"trusted":true},"execution_count":20,"outputs":[{"execution_count":20,"output_type":"execute_result","data":{"text/plain":"(0.935, 0.8766666666666667)"},"metadata":{}}]},{"cell_type":"code","source":"# Saving the vector and the model (Model 4: TfidfVectorizer)\n\nwith open(\"/kaggle/working/tfidf.pkl\", \"wb\") as f:\n    pickle.dump(vect, f)\n    \nwith open(\"/kaggle/working/tfidf_lr.pkl\", \"wb\") as f:\n    pickle.dump(model, f)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:54.010203Z","iopub.execute_input":"2023-08-12T19:26:54.011Z","iopub.status.idle":"2023-08-12T19:26:54.017727Z","shell.execute_reply.started":"2023-08-12T19:26:54.010965Z","shell.execute_reply":"2023-08-12T19:26:54.016663Z"},"trusted":true},"execution_count":21,"outputs":[]},{"cell_type":"markdown","source":"Model 3 (n-grams) performs the best. It performs better than Models 1 (binary), 2 (non-binary) or 4 (Tfidf)!","metadata":{}},{"cell_type":"markdown","source":"### 2 Test Cases (Positive and Negative) with the best model (Model 3: n-grams)","metadata":{}},{"cell_type":"code","source":"# Selecting a positive and a negative review from main table\n\ntest1 = [df[df['Sentiment']==1].iloc[1, 3]]\ntest2 = [df[df['Sentiment']==0].iloc[1, 3]]\n\n# Converting to lower-case\n\ntest1 = [r.lower() for r in test1]\ntest2 = [r.lower() for r in test2]\n\n# Tokenizing words\n\ntest1 = [word_tokenize(r) for r in test1]\ntest2 = [word_tokenize(r) for r in test2]\n\n# Removing stop words\n\ntest1 = [[word for word in t if t not in sw] for t in test1]\ntest2 = [[word for word in t if t not in sw] for t in test2]\n\n# Removing special characters and recombining words\n\ntest1 = [[\" \".join(regex_tokenizer.tokenize(word)) for word in t if len(regex_tokenizer.tokenize(word)) > 0] for t in test1]\ntest2 = [[\" \".join(regex_tokenizer.tokenize(word)) for word in t if len(regex_tokenizer.tokenize(word)) > 0] for t in test2]\n\n# Stemming with the better performing PorterStemmer\n\ntest1 = [[porter.stem(word) for word in t] for t in test1]\ntest2 = [[porter.stem(word) for word in t] for t in test2]\n\n# Recombining words into sentences. This gives the clean text, ready for vectorization and model-fitting.\n\ntest1 = [\" \".join(r) for r in test1]\ntest2 = [\" \".join(r) for r in test2]","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:54.019331Z","iopub.execute_input":"2023-08-12T19:26:54.01974Z","iopub.status.idle":"2023-08-12T19:26:54.044297Z","shell.execute_reply.started":"2023-08-12T19:26:54.019702Z","shell.execute_reply":"2023-08-12T19:26:54.043117Z"},"trusted":true},"execution_count":22,"outputs":[]},{"cell_type":"code","source":"# Loading the best performing Model 3, CountVectorizer with n-gram range 1-3\n\nvect = pickle.load(open(\"/kaggle/working/ngram.pkl\", \"rb\"))\nmodel = pickle.load(open(\"/kaggle/working/ngram_lr.pkl\", \"rb\"))","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:54.046281Z","iopub.execute_input":"2023-08-12T19:26:54.046728Z","iopub.status.idle":"2023-08-12T19:26:54.057913Z","shell.execute_reply.started":"2023-08-12T19:26:54.046686Z","shell.execute_reply":"2023-08-12T19:26:54.056753Z"},"trusted":true},"execution_count":23,"outputs":[]},{"cell_type":"code","source":"# Casting the 2 text reviews as vectors with the best model (Model 3, CountVectorizer (ngrams: 1-3))\nX_test1 = vect.transform(test1)\nX_test2 = vect.transform(test2)\n\n# Generating and validating predictions. The model performs accurately.\n\nmodel.predict_proba(X_test1), model.predict(X_test1), model.predict_proba(X_test2), model.predict(X_test2)","metadata":{"execution":{"iopub.status.busy":"2023-08-12T19:26:54.059223Z","iopub.execute_input":"2023-08-12T19:26:54.059552Z","iopub.status.idle":"2023-08-12T19:26:54.070466Z","shell.execute_reply.started":"2023-08-12T19:26:54.059514Z","shell.execute_reply":"2023-08-12T19:26:54.068763Z"},"trusted":true},"execution_count":24,"outputs":[{"execution_count":24,"output_type":"execute_result","data":{"text/plain":"(array([[0.01462834, 0.98537166]]),\n array([1]),\n array([[9.99954882e-01, 4.51178702e-05]]),\n array([0]))"},"metadata":{}}]},{"cell_type":"markdown","source":"**Consolidated Code**\n\ndf = pd.read_excel('../input/product-reviews/Canva_reviews.xlsx')\n\nreviews = list(df['review'])  \nreviews_lower = [r.lower() for r in reviews]  \ntokens = [word_tokenize(r) for r in reviews_lower]  \n\nsw = stopwords.words('english')  \ntokens = [[word for word in t if word not in sw] for t in tokens]  \n\nporter = PorterStemmer()  \nlancaster = LancasterStemmer()  \n\nporter.stem('teachers'), lancaster.stem('teachers'), porter.stem('absolutely'), lancaster.stem('absolutely')    \n\ntokens = [[porter.stem(word) for word in t] for t in tokens]  \nflat_tokens = [word for t in tokens for word in t]    \n\ncounts = Counter(flat_tokens)  \n\nclean_reviews = [' '.join(t) for t in tokens]\n\nvect = CountVectorizer(binary=True, min_df=5)     \n*change vectorizers to get different models*  \n\nX = vect.fit_transform(clean_reviews)  \nX.shape, len(vect.vocabulary_), type(x)    \n\n\ndf['Sentiment'] = df['Sentiment'].apply(lambda x: 1 if x == 'Positive' else 0)   \n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42)  \n\nX_train.shape, X_test.shape, y_train.shape, y_test.shape  \n\n\nmodel = LogisticRegression()  \nmodel.fit(X_train, y_train)  \ny_pred = model.predict(X_test)  \naccuracy_score(y_test, y_pred)  ","metadata":{}}]}