├── README.md
└── demo.py


/README.md:
--------------------------------------------------------------------------------
 1 | # twitter_sentiment_challenge
 2 | Twitter Sentiment Analysis Challenge for Learn Python for Data Science #2 by @Sirajology on Youtube
 3 | 
 4 | ##Overview
 5 | 
 6 | This is the code for the Twitter Sentiment Analyzer challenge for 'Learn Python for Data Science #2' by @Sirajology on [YouTube](https://youtu.be/o_OZdbCzHUA). The code uses the [tweepy](http://www.tweepy.org/)  library to access the Twitter API and the [TextBlob](https://textblob.readthedocs.io/en/dev/) library to perform Sentiment Analysis on each Tweet. We'll be able to see how positive or negative each tweet is about whatever topic we choose. 
 7 | 
 8 | ##Dependencies
 9 | 
10 | * tweepy (http://www.tweepy.org/)
11 | * textblob (https://textblob.readthedocs.io/en/dev/)
12 | 
13 | Install missing dependencies using [pip](https://pip.pypa.io/en/stable/installing/)
14 | 
15 | ##Usage
16 | 
17 | Once you have your dependencies installed via pip, run the script in terminal via
18 | 
19 | ```
20 | python demo.py
21 | ```
22 | 
23 | ##Challenge
24 | 
25 | Instead of printing out each tweet, save each Tweet to a CSV file with an associated label. The label should be either 'Positive' or 'Negative'. You can define the sentiment polarity threshold yourself, whatever you think constitutes a tweet being positive/negative. Push your code repository to [github](https://help.github.com/articles/set-up-git/) then post it in the comments. I'll give the winner a shoutout a week from now!
26 | 
27 | ##Credits
28 | 
29 | This code is 100% Siraj baby.
30 | 


--------------------------------------------------------------------------------
/demo.py:
--------------------------------------------------------------------------------
 1 | import tweepy
 2 | from textblob import TextBlob
 3 | 
 4 | # Step 1 - Authenticate
 5 | consumer_key= 'CONSUMER_KEY_HERE'
 6 | consumer_secret= 'CONSUMER_SECRET_HERE'
 7 | 
 8 | access_token='ACCESS_TOKEN_HERE'
 9 | access_token_secret='ACCESS_TOKEN_SECRET_HERE'
10 | 
11 | auth = tweepy.OAuthHandler(consumer_key, consumer_secret)
12 | auth.set_access_token(access_token, access_token_secret)
13 | 
14 | api = tweepy.API(auth)
15 | 
16 | #Step 3 - Retrieve Tweets
17 | public_tweets = api.search('Trump')
18 | 
19 | 
20 | 
21 | #CHALLENGE - Instead of printing out each tweet, save each Tweet to a CSV file
22 | #and label each one as either 'positive' or 'negative', depending on the sentiment 
23 | #You can decide the sentiment polarity threshold yourself
24 | 
25 | 
26 | for tweet in public_tweets:
27 |     print(tweet.text)
28 |     
29 |     #Step 4 Perform Sentiment Analysis on Tweets
30 |     analysis = TextBlob(tweet.text)
31 |     print(analysis.sentiment)
32 |     print("")
33 | 


--------------------------------------------------------------------------------