Monitoring and Exporting of live tweets of twitter users and doing Sentiment Analysis on them.
This application basically consists of 3 main areas to focus:
-
Monitoring : Real time streaming of tweets on screen and it's respective sentimental value
-
Downloading / Exporting : Exporting the tweet data with details including
- Tweet created_at,
- User's screen_name,
- language of tweet,
- Tweet text,
- UserProfileName,
- UserDescription,
- UserLocation,
- FavoritesCount,
- TotalTweetsCount,
- UserFriendCount,
- UserFollowersCount,
- TimeZone,
- UTCOffset Value,
- GeoEnabled,
- IfUserVerified,
- IfContributors,
- TweetSource,
- coordinates,
- SentimentPolarity,
- SentimentSubjectivity
Here we are using SQLite3 database to export these details.
- Sentimental Analysis: We are using module 'Textblob' to do sentimental analysis. In this module, sentimental value is further categorized into two sub-parts SentimentPolarity and SentimentSubjectivity. The value of SentimentPolarity range from -1 to +1.
- -1 indicates sentiments of give text is Negative
- +1 indicates sentiments of give text is Positive
- 0 indicates sentiments of give text is Neutral
Future room for upgrade / enhancements:
- Can improve analysis part.