Public Mood

Know public sentiment towards one topic in real time! Play with it and find interesting stuff of most recent trends.

In the welcome page, user can search any keyword, then the app will automatically get at most 100 relevant tweets from twitter from most recently. Then the app will analysis the tweets using a self-trained machine learning model, to identify each tweet as positive, negative or neutral.

The second page will show the proportions of 'positive', 'negative', 'neutral', and show an emoji of the overall public sentiment, whether is very happy of the keyword, or other sentiment. The app will also generate a word cloud of the top 200 used words, the higher the frequency one word appears in tweets, the larger the word will be in the word cloud. Then, if user are curious of what these tweets actually are, he/she can tap the 'click for details' button, the third page will appear.

The details page will show all the tweets details with user avatar, user's name, location, and full text tweet.

Tapping ' < ' will direct the user back to the second page, and tapping ' < ' in the second page will direct user back to the first welcome page, to try different keywords and see what people think about it in twitter.