What is this?

• data mining using tweepy (search by key words ex: 'trump')
• sentiment analysis using NLTK, textblob etc
• words frequence count from a batch of tweeter messages (remove stopwords and stemming)
• calling methods by API (ex: localhost:5000/search?query=trump&tweet_limit=100&word_freq=20)

TODO:

1. mysql for storeage
2. front-end page
3. get all tweets from one user's
4. proper auth

Mining Twitter Data with python + mysql!

---

To run you need:

1. python 3.7.0
2. MySQL
3. Twitter Account with consumer key and access token (get approved from twitter first)

Example of app_config.py file

OAuth Authentication

consumer_key=""
consumer_secret=""
access_token=""
access_token_secret=""

Database

HOST = "127.0.0.1"
USER = "root"
PASSWD = "mypwd"
DATABASE = "twt_database"

Installation

1. we need tweepy
2. we need TextBlob for sentiment analysis (note that NLTK download is about 3.5 GB)

pip install tweepy
pip install nltk
pip install textblob

$ python
>>> import nltk
>>> nltk.download()

We have to select “all” and click download.