otifsolutions/php-sentiment-analysis

Php Sentiment Analysis allows user to analyze sentiments from a text statement

v1.0.2 2020-06-04 08:59 UTC

This package is auto-updated.

Last update: 2024-05-09 12:03:23 UTC


README

Install via Composer

Using Composer (Recommended)

You can run the following command in the root directory of your project:

composer require otifsolutions/php-sentiment-analysis

Usage

  • Php Sentiment Analysis allows to analyze a text statement if it is positive or negative. e.g. comment on any post.

    Sentiment::make()->analyze($value);
    • Where $value is the statement to be analyzed.
  • Statement may contain labels, emojis and negators.

    • Labels are words statement contains e.g. corruption, boycott.
    • Emoji e.g. 😍, 💖.
    • Negators are words that negate something e.g. doesnt, not.
  • We use following methods

    • getTokens

      • This method separate the words from the given statement and returns all words.
    • getDataSet

      • This method has dataset.
      • Labels, emojis and negators already stored in the dataset, which have a positive or negative value.
      • For example "fun": 4, "funeral": -1,

This example demonstrates the usage of Php Sentiment Analysis,

  Sentiment::make()->analyze('She is famous, but she hates others.');
  {
   "score": -1,
   "comparative": -0.14285714285714285,
   
   "calculation": [
     {"she": 0},{"is": 0},{"famous": 2},{"but": 0},{"she": 0},{"hates": -3},{"others": 0}
     ],
  
   "tokens":  [
    "she","is","famous","but","she","hates","others"
      ],
 
   "positive": [
      "famous"
      ],
 
   "negative": [
      "hates"
      ]
  }
Option Description
score Positive or negative value after analysis of given statement.
tokens Tokens are the separated words in a Statement.
comparative Value received when score is divided by count of token.
calculation Words in an array with a positive or negative value, if the word exists in dataset the value will be positive or negative otherwise value will be zero.
positive Shows all positive words from the statement.
negative shows all negative words from the statement.