merelinguist/mindreader

Super simple sentiment analysis with AFINN.

Super simple sentiment analysis with AFINN.

Usage

npm i styled-preflight

yarn add styled-preflight

import { mindreader } from 'mindreader';

const sentiment = mindreader('good, bad');

console.log(sentiment);

/*
{
  score: -3,
  comparative: -1,
  calculation: [{ ugly: -3 }, { bad: -3 }, { good: 3 }],
  tokens: ['good', 'bad', 'ugly'],
  words: ['ugly', 'bad', 'good],
  positive: ['good']
  negative: ['ugly', 'bad']
}
*/

Input & Ouput

Input

mindreader takes any string input, for example:

import { mindreader } from 'mindreader';

const sentiment = mindreader(
  'It was the best of times, it was the worst of times...'
);

...

Output

The full output looks like this:

Name	Type	Description
score	number	The overall sentiment score. Essentially just the sentiment ratings of each token, added up.
comparative	number	If you want to compare several strings, use this. It calculates the average score of all the tokens in your provided string.
calculation	{ [key: string]: number }[]	How did we get here? This outputs the score which was assigned to each individual token in the string. If you want to do word-by-word sentiment analysis, passing in a big string to mindreader, and then reading from caluclation is the fastest way of doing it.
tokens	string[]	A list of the words in your string.
words	string[]	All the words which mindreader actually scanned. Some stopwords, like “the”, or rarer words will not be included here.
positive	number	A list of the positive marked words (those scored above 0) .
negative	number	A list of the positive marked words (those scored below 0) .

Prior art

• https://github.com/thisandagain/sentiment