An iOS app to tell us about the current sentiment of a keyword on Twitter.
- I have used the dataset from Kaggle website (https://www.kaggle.com/crowdflower/twitter-airline-sentiment) to train my model using CreateML.
- The I imported the ml model into my Xcode project.
- The user types a keyword for which he wants to know the current sentiment. The latest tweets containing that keyword are fetched from Twitter using its API.
- Then all the tweets are classified into 3 categories: positive, negative and neutral using CoreML. Based on the number of tweets in each category, an emoji is displayed to the user.
Screenshots from the app: