https://github.com/mtrebi/sentimentanalyzer

Sentiment Analyzer with RapidMiner
https://github.com/mtrebi/sentimentanalyzer

data-mining filmaffinity films-reviews prediction rapidminer review sentiment-analysis

Last synced: 6 days ago
JSON representation

Sentiment Analyzer with RapidMiner

• Host: GitHub
• URL: https://github.com/mtrebi/sentimentanalyzer
• Owner: mtrebi
• License: mit
• Created: 2016-02-14T10:25:56.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2016-02-14T10:46:06.000Z (over 8 years ago)
• Last Synced: 2023-03-01T08:05:45.630Z (over 1 year ago)
• Topics: data-mining, filmaffinity, films-reviews, prediction, rapidminer, review, sentiment-analysis
• Size: 3.79 MB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# SentimentAnalyzer

Sentiment Analyzer with RapidMiner that

- Collects latest film reviews from a website (scraping)

- Process the reviews. Tokenizing, removing stopwords and outliers and stemming.

- Analyze data and generated a Naive Bayes and a Support Vector Machine model.

- Using the model generated in the last step, predicts the sentiment of new films reviews.

### Data

Holds part of data that will be processed

The **pang_lee** folder contains the training and test data

The **filmaffinity** folder contains manually downloaded data to do predictions

### Process

Contains all the RapidMiner files.

The **generator** process takes the data from the data folder and trains and tests a model

The **optimizer** process iterates the same model with differents parameters to find the best one

The **scraper** process downloads new film reviews from www.rottentomatoes.com to do predictions

The **predictor_rottentomatoes** does predictions to the downloaded film reviews from that page

The **predictor_filmaffinity** does predictions to the manually downloaded film reviews from the filmaffinity data folder