Sentiment Parser

Description

An npm Package for Node.js to analyze the content of comments based on categories of words.

To run the demo, clone the repo and run

$ node demo.js 

Basic Usage

const { parseComments } = require("sentiments-parser")

const comments = [
"This is nice work :)",
"Please stop!",
"Delightful =) Adore the use of gradient and background!",
"Incredibly thought out! Wow love it!",
"Such sleek.",
"It's fun not just splendid!",
"Just fresh."]

const categories = {
"Selfcentered" : ["i", "me", "classic", "...", "simple"],
"Smileys" : [":)", "=)", ":-)"],
"Douche" : ["dude", "mate", "m8", "fresh", "laugh"],
"Positive" : ["nice", "delightful", "love", "sleek", "fun", "clean", "appealing", "like"],
"Extra positive": ["inspiring", "excellent", "splendid", "gorgeous"],
"Over the top": ["heart", "tremendously", "fab", "blew", "mind", "mission"],
"Tech talk": ["gradient", "background", "style", "minimal", "lines", "navigation", "colour", "palette", "boldness", "camera", "angle", "image", "avatar", "concept", "shot", "cyan", "shapes"]
}


parseComments(comments, categories)

Example Outcome

The score consist in the percentage of the amount of space the occurring words take up in a comment

   {
     comment: "Delightful =) Adore the use of gradient and background!",
    "Selfcentered" : {score: 0, words: []},
    "Smileys" : {score: "3.64", words: ["=)"]},
    "Douche" : {score: 0, words: []},
    "Positive" : {score: "18.18", words: ["delightful"]},
    "Extra positive": {score: 0, words: []},
    "Over the top": {score: 0, words: []},
    "Tech talk": {score: "32.73", words: ["gradient", "background"]}  
   }