1.1.2 • Published 6 years ago

empath-sentiment-analysis v1.1.2

Weekly downloads
1
License
ISC
Repository
github
Last release
6 years ago

Empath lets you perform client side sentiment analysis. It is designed to work with emails.

Features include:

  • multilingual sentiment analysis (German and English supported currently)
  • DISC profile analysis
  • email signature recognition (so only the email body gets analysed)
  • egoismAnalyser (so you can figure out how selfish someone is based on their communication)

Usage:

Install module

npm install --save-dev empath-sentiment-analysis

Using Sentiment Analyser

Modified from https://github.com/thisandagain/sentiment/blob/master/lib/index.js

import empath from 'empath-sentiment-analysis';
empath.analyseSentiment(phrase = '', attributesOfInterest = []);

Result:

{
    wordListDidNotMatch: tryAnotherWordList,
    score:          score,
    comparative:    score / tokens.length,
    tokens:         tokens,
    words:          words,
    positive:       positive,
    negative:       negative,
    attributeScores: attributeScores
};

Attribute scores are useful if you have a number of products for which you want to analyse sentiment. Review the test cases for more details.

Using DISC Profile Guesser

import empath from 'empath-sentiment-analysis';
empath.guessDISCProfile.getUserReadableDISCProfile(email);

result:

{
    'D': D,
    'I': I,
    'S': S,
    'C': C,
}

or empath.guessDISCProfile.getUserReadableDISCProfile(email);

result: two letters indicating primary and secondary profile, e.g. DC

Using Email parser

import empath from 'empath-sentiment-analysis';
const emailWithoutSignatureAndQuotedReplies = empath.parseEmail(email);

If you supply an email with a long signature and quoted replies from the thread, it will trim everything except the first email.

Using Readability analyser

import empath from 'empath-sentiment-analysis';
const readabilityScore = empath.calculateReadabilityScore(email);

Anything above 4.0 is considered advanced vocab

Egoism analyser

import empath from 'empath-sentiment-analysis';
const egoismScore = empath.analyseEgoism(email);

Result:

{
    selfish, 
    controlling, 
    conforming
};

Selfish = number of selfish words used Controlling = number of command/imperative words used Conforming = number of we/group style words

Ratio can yield interesting insights about personality

1.1.1

6 years ago

1.1.0

6 years ago

1.1.2

6 years ago

1.0.7

10 years ago

1.0.6

10 years ago

1.0.5

10 years ago

1.0.4

10 years ago

1.0.3

10 years ago

1.0.2

10 years ago

1.0.1

10 years ago

1.0.0

10 years ago