Wondering what’s next for npm?Check out our public roadmap! »

    ml-sentiment

    2.0.7 • Public • Published

    npm version build status Dependency Status devDependency Status Code Coverage Downloads Today Downloads Month

    Wat

    Simple text sentiment analyser.

    Install

    npm install ml-sentiment

    Usage

    var ml = require('ml-sentiment')
    ml.classify('Rainy day but still in a good mood')
    //=> 2 ... (overall positive sentiment)

    How

    Returns a positive number for positive sentiment association and negative number for negative sentiment association.

    Basics

    var longSentence = `Transform json to csv data. The difference to my other
    module json2csv is json2csv-stream uses streams for transforming the incoming
    data. The module is built with the new streaming API from Node.js v0.10.0 but
    maintains backwards compatibility to earlier Node.js versions. Listen for
    header and line events or pipe the data directly to a readable stream.`
     
    var ml = require('ml-sentiment')()
    ml.classify(longSentence)
    //=> 0 ... (very boring encyclopedia like text)
     
    ml.classify('Rainy day but still in a good mood')
    //=> 2 ... (overall positive sentiment)

    Negations

    var ml = require('ml-sentiment')()
    ml.classify(`not awesome`)
    //=> -3 (negative)
     
    ml.classify(`awesome`)
    //=> 3 (positive)

    German

    var ml = require('ml-sentiment')({lang: 'de'})
    ml.classify(`Es ist nicht so toll`)
    //=> (negative)

    Credits

    Original model and data: Finn Årup Nielsen, "A new ANEW: Evaluation of a word list for sentiment analysis in microblogs", http://arxiv.org/abs/1103.2903

    For german model: R. Remus, U. Quasthoff & G. Heyer: SentiWS - a Publicly Available German-language Resource for Sentiment Analysis. In: Proceedings of the 7th International Language Ressources and Evaluation (LREC'10), 2010

    Install

    npm i ml-sentiment

    DownloadsWeekly Downloads

    222

    Version

    2.0.7

    License

    ISC

    Last publish

    Collaborators

    • avatar