cat README.md

Liberal Hate Machine

The Liberal Hate Machine is a project that attaches to a stream of tweets targeted at President Donald J. Trump and Gab.com. It is performing real-time matching on the text of the tweets looking for a list of keywords and phrases commonly used online in Tweets expressing outrage and hate towards the President and conservatives.

This project was created in response to Federal Funds Used To Research 'Hate Speech' On Gab, and is now operated as a real-time analysis of hate on Twitter.com.

The system is saving every tweet it encounters, but marking each one with a matched keyword or phrase as triggered. The triggered tweets are then used to build the home page of this site. Search lets you explore all tweets recorded by this system. This is not all tweets in the world. It is only the tweets sent by @realDonaldTrump and @getongab as well as all replies and mentions sent by the general public to those accounts.

Speech Classification

The system is doing basic keyword matching as well as general keyword sentiment analysis. Liberal Hate Machine is using the sentiment Node.js library to process each tweet as they arrive and generate a score as well as some other more detailed analysis. Every tweet in our database records the following additional computed metadata:

{
    "score": 1,
    "comparative": 0.1111111111111111,
    "calculation": [ { "allergic": -2 }, { "love": 3 } ],
    "tokens": [
        "i",
        "love",
        "cats",
        "but",
        "i",
        "am",
        "allergic",
        "to",
        "them"
    ],
    "words": [
        "allergic",
        "love"
    ],
    "positive": [
        "love"
    ],
    "negative": [
        "allergic"
    ]
}

This additional metadata is then used to help populate the home page with recent tweets that score poorly as well as provide you with additional metrics to use with Search such as limiting to a specific score range.


Copyright © 2020 Gab AI, Inc.
Terms of Service