Final Project - Machine Learning

Name: Kfir Bar

Title: Sentiment Analysis of Movie Reviews and Twitter Statuses

Main project report

Datasets

Name Reference After preprocessing
IMDB - movie reviews web site NA (I worked directly on the source files)
Twitter web site Twitter data file

Source code

File name Description
TwitterApplication.java This file implements the entire flow of the experiments we performed for this project (both datasets)
OpenNLPRunner.java This class wraps the capabilities of the OPENNLP library for processing English texts. There is also an integration here with WordNet for determining the lemma of a given word
config The configuration file for our application
emoticons Containing a list of emoticons that we consider as words in our algorithms

Accuracy results file

References to open source projects that I used

OPEN-NLP
Weka
WordNet