mr4jay/PyTextSentiment

Emotion Detection & Classification of Tweets Using Streaming APIs. [NLTK] [Scikit Learn] [Twitter Streaming API] [Bing API]

Python

Emotion Detection And Classification of Tweets

How to Run

Go to your terminal.

Clone this repository and go inside it

 git clone https://github.com/mjaglan/PyTextSentiment.git
 cd PyTextSentiment

Edit following files:

 app/assets/BingCredentials/bingClientId.txt
 app/assets/BingCredentials/bingClientSecret.txt
 app/assets/TwitterAPI/credentials.txt
 app/assets/input.txt

Run the following script
```
 . ./restart-all.sh
```

Web UI

This application inside docker container page can be browsed at http://CONTAINER-IP:9091/

Project Structure

Training dataset files: app/assets/Resource/searchKeys
Get Feeds by text search query: app/assets/Resource/searchKeys/testFiles
Bag of Words for emotion tagging and classification: app/assets/Resource
Output twitter data directory: app/assets/twitterData

Project Overview

The term paper of this work is present here. Below are the highlights of the work done + the results generated for live tweets on 28 December 2015.

References