twitter, tweet, sentiment, android
Pull request Compare This branch is even with masonwan:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.
Android
classifier
.gitignore
README.md

README.md

Project Title

Tweentiment

Search screen Result screen

Introduction

Twitter is a social network and microblogging system which limits a single content to 140 words. This limitation did not hinder the grows of users but rather attract people or organization to share, advertise or interact on the platform. 50 millions tweets was sent per day in average. This project is building an Android app which displays and calculates the sentiment value for each tweet. The goal is to achieve a competitive result with Sentiment140 API.

This application is using Twitter Search API. The json files are parsed using GSON library.

How to run

You must install

Use Eclipse to open each .profile file in Android and Classifier directories and build project. For evaluator, the result will be saved into a JSON file in the working directories.

Related Work

Several websites have already implemented this idea. The top two from Google search results are Sentiment140 and TweetFeel.

Evaluation

Compare our result with Sentiment140 result (API available). List the differences and let human judge which one is more accurate.

Evaluator

References

Definitions

Sentiment analysis - Wikipedia, the free encyclopedia http://en.wikipedia.org/wiki/Sentiment_analysis

Dataset

A new ANEW: Evaluation of a word list for sentiment analysis in microblogs http://arxiv.org/abs/1103.2903

Projects

Twitter Sentiment Analysis Tutorial Datasets, papers, presentations https://github.com/jeffreybreen/twitter-sentiment-analysis-tutorial-201107

Sentiment - Nodejs implementation for text sentiment analysis Very simply; use AFINN as classifier https://github.com/thisandagain/sentiment

Troll - Node.js and neural network implementation for sentiment analysis Need Redis, basically no training data https://github.com/thisandagain/troll

Twitter sentiment analysis using Python and NLTK | Laurent Luce's Blog http://www.laurentluce.com/posts/twitter-sentiment-analysis-using-python-and-nltk/

Studies

Effective Listings of Function Stop words for Twitter Finding Twitter Stop Words http://arxiv.org/abs/1205.6396

Twitter Sentiment Analysis - AFINN http://fnielsen.posterous.com/afinn-a-new-word-list-for-sentiment-analysis

Opinion mining from noisy text data http://dl.acm.org.libaccess.sjlibrary.org/citation.cfm?id=1390763&dl=GUIDE&coll=GUIDE

CyberEmotions - A large-scale collective emotion research for e-communities. http://en.wikipedia.org/wiki/CyberEmotions

Mining and summarizing customer reviews http://dl.acm.org/citation.cfm?id=1014073

Opinion observer http://dl.acm.org/citation.cfm?id=1060797

Sentiment Analysis and Subjectivity A comprehensive tutorial for text sentiment analysis methods. It’s highly recommended to read. http://dl.acm.org/citation.cfm?id=1218990