Request - mode for infering single article topics

[ilanle]

Hello again,
Let's say that we've gone through the corpus and created topic word distributions.
I want to use this output to tag single articles now.
I know that the process is similar, iterative, only that it needs not to affect the phi.
I think such a function would be useful for people like myself.
Thanks
Ilan

[primaryobjects]

Maybe a separate node module that builds upon lda?

[matnel]

Any updates on this? Is there a module I should use to conduct this?

[primaryobjects]

Not sure, do you have an example? This might be something that could be a separate library, built on top of lda.

[yawetse]

Hi Kory, thanks for the amazing module. Would it still be possible to add this functionality within the same module?

The main reason is to convert text documents into feature vectors for a supervised learning problem.

Something like this (modifed documentation example):

const lda = require('lda');
 
// Example document.
const text = 'Cats are small. Dogs are big. Cats like to chase mice. Dogs like to eat bones.';
 
// Extract sentences.
const documents = text.match( /[^\.!\?]+[\.!\?]+/g );
 
// Run LDA to get terms for 2 topics (5 terms each).
const LDAModel = lda(documents, 2, 5);

/*
LDAModel.results = [ [ { term: 'dogs', probability: 0.2 },
    { term: 'cats', probability: 0.2 },
    { term: 'small', probability: 0.1 },
    { term: 'mice', probability: 0.1 },
    { term: 'chase', probability: 0.1 } ],
  [ { term: 'dogs', probability: 0.2 },
    { term: 'cats', probability: 0.2 },
    { term: 'bones', probability: 0.11 },
    { term: 'eat', probability: 0.1 },
    { term: 'big', probability: 0.099 } ] ];
*/

//and hopefully a predict method on a trained model
LDAModel.predict('cats are big') // => [0.89,0.3] (calculate theta for new documents)

Again thanks for the great module @primaryobjects!