Emojis in dataset for Sentiment Analysis

Question

The Learning Path for sentiment analysis says that it can evaluate emojis.  I want to create a dataset that contains emojis.  What type of file and what encoding do I need to use so that the Sentiment analysis tool will take a 😀 as a positive sentiment.   Also wondering if there is anything that tells us which emojis can be evaluated?

Thanks.

echuong1 · Answer

Try taking a look at the lexicon:

https://github.com/cjhutto/vaderSentiment/blob/master/vaderSentiment/vader_lexicon.txt