Nectar of the Programming Masses

wink-naive-bayes-text-classifier

1.1.4 • Public • Published

wink-naive-bayes-text-classifier

Configurable Naive Bayes Classifier for text with cross-validation support

Build Status Coverage Status Inline docs dependencies Status devDependencies Status

Classify text, analyse sentiments, recognize user intents for chatbot using wink-naive-bayes-text-classifier. It is a part of wink — a growing family of high quality packages for Statistical Analysis, Natural Language Processing and Machine Learning in NodeJS.

It's API offers a rich set of features:

  1. Configure text preparation task such as amplify negation, tokenize, stem, remove stop words, and propagate negation using wink-nlp-utils or any other package of your choice.
  2. Configure Lidstone or Lapalce additive smoothing.
  3. Configure Multinomial or Binarized Multinomial Naive Bayes model.
  4. Export and import learnings in JSON format that can be easily saved on hard-disk.
  5. Evaluate learning to perform n-fold cross validation.
  6. Obtain comprehensive metrics including confusion matrix, precision, and recall.

Installation

Use npm to install:

npm install wink-naive-bayes-text-classifier --save

Example

 
// Load Naive Bayes Text Classifier
var Classifier = require( 'wink-naive-bayes-text-classifier' );
// Instantiate
var nbc = Classifier();
// Load NLP utilities
var nlp = require( 'wink-nlp-utils' );
// Configure preparation tasks
nbc.definePrepTasks( [
  // Simple tokenizer
  nlp.string.tokenize0,
  // Common Stop Words Remover
  nlp.tokens.removeWords,
  // Stemmer to obtain base word
  nlp.tokens.stem
] );
// Configure behavior
nbc.defineConfig( { considerOnlyPresence: true, smoothingFactor: 0.5 } );
// Train!
nbc.learn( 'I want to prepay my loan', 'prepay' );
nbc.learn( 'I want to close my loan', 'prepay' );
nbc.learn( 'I want to foreclose my loan', 'prepay' );
nbc.learn( 'I would like to pay the loan balance', 'prepay' );
 
nbc.learn( 'I would like to borrow money to buy a vehicle', 'autoloan' );
nbc.learn( 'I need loan for car', 'autoloan' );
nbc.learn( 'I need loan for a new vehicle', 'autoloan' );
nbc.learn( 'I need loan for a new mobike', 'autoloan' );
nbc.learn( 'I need money for a new car', 'autoloan' );
// Consolidate all the training!!
nbc.consolidate();
// Start predicting...
console.log( nbc.predict( 'I would like to borrow 50000 to buy a new Audi R8 in New York' ) );
// -> autoloan
console.log( nbc.predict( 'I want to pay my car loan early' ) );
// -> prepay
 

API

definePrepTasks( tasks )

Defines the text preparation tasks to transform raw incoming text into an array of tokens required during learn(), evaluate() and predict() operations. The tasks should be an array of functions. The first function in this array must accept a string as input; and the last function must return an array of tokens as JavaScript Strings. Each function must accept one input argument and return a single value. definePrepTasks returns the count of tasks.

As illustrated in the usage, wink-nlp-utils offers a rich set of such functions.

defineConfig( config )

Defines the configuration from the config object. This object must define 2 properties viz. (a) considerOnlyPresence and smoothingFactor. The considerOnlyPresence must be a boolean — true indicates a binarized model; default value is false. The smoothingFactor defines the value for additive smoothing; its default value is 1. The defineConfig() must be called before attempting to learn.

learn( input, label )

Simply learns that the input belongs to the label. If the input is a JavaScript String, then definePrepTasks() must be called before learning.

consolidate()

Consolidates the learning. It is a prerequisite for evaluate() and/or predict().

evaluate( input, label )

It is used to evaluate the learning against a test data set. The input is used to predict the label, which is compared with the label to populate a confusion matrix.

metrics()

It computes a detailed metrics consisting of macro-averaged precision, recall and f-measure along with their label-wise values and the confusion matrix.

predict( input )

Predicts the label for the input. If it is unable to predict then it returns a value 'unknown'.

computeOdds( input )

Computes the log base-2 of odds of every label for the input; and returns the array of [ label, odds ] in descending order of odds. Here is an example of the returned array:

[
  [ 'prepay', 6.169686751688911 ],
  [ 'autoloan', -6.169686751688911 ]
]

If it is unable to make prediction then it returns a value [ [ 'unknown', 0 ] ].

exportJSON()

The learning can be exported as JSON text that may be saved in a file.

importJSON( json )

An existing JSON learning can be imported for prediction. It is essential to definePrepTasks() and consolidate() before attempting to predict.

stats()

Returns basic stats of learning in terms of count of samples under each label, total words, and the size of vocabulary.

reset()

It completely resets the classifier by re-initializing all the learning related variables, except the preparatory tasks. It is useful during cross fold-validation.

Need Help?

If you spot a bug and the same has not yet been reported, raise a new issue or consider fixing it and sending a pull request.

Copyright & License

wink-naive-bayes-text-classifier is copyright 2017 GRAYPE Systems Private Limited.

It is licensed under the under the terms of the GNU Affero General Public License as published by the Free Software Foundation, version 3 of the License.

install

npm i wink-naive-bayes-text-classifier

Downloadsweekly downloads

17

version

1.1.4

license

AGPL-3.0

homepage

github.com

repository

Gitgithub

last publish

collaborators

  • avatar
  • avatar
  • avatar
Report a vulnerability