Word Concurrence

Word Concurrence

The goal of this project is to find statistics about word pairs in tweets. It also uses the AFINN file, which specifies the sentiment of common words. We can then find out what people are talking about, how often they do and whether they are positive things.

Usage

1. Get the scores of each word and find pairs

$ get_concurrence [afinnFile] [jsonTweetsFile] [destinationDirectory]

Explanation

1. The program uses the AFINN file to sum up the score of a word each time it is found in a tweet
2. The score is recorded into score.csv
3. Then for each pair of the words found together in the same tweet, the concurrence count and the word-to-word-pair ratio is recorded
4. The results are exported to the file concurrence