Package Exports

term-frequency

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (term-frequency) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

term-frequency

A simple term frequency library that takes in a document vector, and compiles the frequency calculation of your choosing.

First make the necessary require-ments

var sw = require('stopword')
var tf = require('term-frequency');
var tv = require('term-vector');

You can then do:

var vec = tv.getVector(
  sw.removeStopwords(
    'This is a really, really cool vector. I like this VeCTor'
      .toLowerCase()
      .split(/[ ,\.]+/)
  )
)
var freq = tf.getTermFrequency(vec);
// freq is now
// [ [ [ 'cool' ], 1 ], [ [ 'really' ], 2 ], [ [ 'vector' ], 2 ] ];

Or you can specify a TF scheme like so:

var vec = tv.getVector('This is a really, really cool vector. I like this VeCTor');
var freq = tf.getTermFrequency(vec, {scheme: tf.logNormalization});
// freq is now:
// [
//   [ [ 'cool' ], 0.6931471805599453 ],
//   [ [ 'really' ], 1.0986122886681098 ],
//   [ [ 'vector' ], 1.0986122886681098 ]
// ]);

Currently supported schemes are

raw
logNormalization
doubleNormalization0point5
selfString
selfNumeric

See the Wikipedia page for more info about term frequency calculation

You can also weight your calculations like so. A weight is a numeric value that will be added to the calculated score.

var freq = tf.getTermFrequency(vec, {
  scheme: tf.doubleNormalization0point5, 
  weight: 5
});
// freq is now
// [
//   [ [ 'cool' ], 5.7027325540540822 ],
//   [ [ 'really' ], 5.9581453659370776 ],
//   [ [ 'vector' ], 5.9581453659370776 ] 
// ]);