lexicon-sentiment-analysis

Simple example of using a lexicon to perform sentiment analysis

Usage

python ./doAnalysis.py

The lexicon

Included in this repository is a lexicon is the MPQA subjectivity lexicon. See the file subjectivity_clues_hltemnlp05/subjclueslen1-HLTEMNLP05.README for more information.

I've created a simpler format of the lexicon, using the subjectivity_clues_hltemnlp05/reformat.py script. The resulting file is ``subjectivity_clues_hltemnlp05/lexicon_easy.csv, and that's what the doAnalysis.py` script uses.

Data

Included in this repository is a dataset of tweets. Each row contains three columns:

1. The tweet text.
2. The tweet ID.
3. The tweet publish date

The full dataset (i.e., newtwitter.csv contains 8,595 rows. A small sample of 100 rows (i.e., newtwitter.small.csv) is also provided.