public-transport-sentiment-lists

This repository provides a list of phrases, which can be used to evaluate the text is a positive or negative statement. Its heavy inspired by AFINN lists but uses complete phrase instead of single words.

current provided lists are:

• de_de

If you what to use this list, you can use the the project sentiment-custom-lists which is written in javascript

Contribution

rate tweets and add phrases

There is automatic process in the background which adds new relevant tweets to the assets. If you want to help, create a pull-request and rate the automatic added tweets and/or add phrases to the lists

Add a phrase

Format:

[REGULAR_EXPRESSION]<TAB>[RATING]

Example:

(zug|bus) fällt aus -1

run the tests:

npm install
npm test

I know a twitter account which can provide test data

Please contact me. I can add the account to the process