A small project from my Natural Language Processing Course, based on Blair-Goldensohn et al.'s "Building a Sentiment Summarizer for Local Service Reviews" (2008). This project extracts aspects using part of speech tags and aggregates sentiment scores across the review. The technique to score sentiments is Unsupervised, based on the works of Opinion Mining, Sentiment Analysis, and Opinion Spam Detection (Bing Liu). The demo_one.py file lets you to hardcode a review and score sentiments.
dsaam94/opinion_mining
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|