Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    Hi @niranja33050187_twitter yes, it is language agnostic.
    niranjan
    @niranja33050187_twitter
    Can I see an example for language support using wink-naive-bayes-text-classifier?
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    @niranja33050187_twitter please checkout https://runkit.com/sanjaya/6128e6f8192462001548b330 at RunKit.
    niranjan
    @niranja33050187_twitter
    @sanjayaksaxena thanks for the example. really helpful library. Also just fr understanding sake, does the library use Python internally with Javascript as wrapper on it?
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    @niranja33050187_twitter All winkjs packages, including wink-naive-bayes-text-classifier, are pure javascript and have zero external dependencies.
    1 reply
    Phill Duffy
    @phillduffy
    Hi - is it possible to detect questions? It's proving a difficult thing to Google for.
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    @phillduffy please provide more context so that one can understand the exact question.
    Eric Vida
    @ericvida
    How can I make new models for new languages. Is there a repo with different language models?
    Eric Vida
    @ericvida
    I don't see any good documentation on how to make models such as the existing english model.
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    Hi @ericvida right now the model creation part is not yet open-sourced, that is why there is no documentation. We are working towards making it ready for o/s.
    Stellasphere
    @stellasphere
    Is it possible to do custom named entity recognition with winkjs?
    Rachna Chakraborty
    @rachnachakraborty
    Hi @stellasphere yes winkNLP has custom entity definition and recognition. Follow this link to get started https://winkjs.org/wink-nlp/custom-entities.html
    Stellasphere
    @stellasphere
    Hi @rachnachakraborty thanks for the quick response. I'll check more in-depth later, but at first glance, it seems like this is POS tagging.
    I may be wrong about it as I am still new to NLP, but I am trying to recreate what MonkeyLearn's "Text Extractor" does (https://monkeylearn.com/text-extractors/) which I think is called Named Entity Recognition. Is that what winkjs's custom entities does?
    Rachna Chakraborty
    @rachnachakraborty
    wink-nlp has APIs for extracting Entities ( Named Entity Recognition ) and Custom Entities as well. You can define text patterns comprising of Postags, Named Entities, Literals and much more. Please refer to this link for defining custom entities https://winkjs.org/wink-nlp/learn-custom-entities.html
    Saurav Joshi
    @sauravjoshi23
    image.png
    Hi, can anyone please share the slack workspace id which can be entered here?
    Saurav Joshi
    @sauravjoshi23
    As I want to join the slack channel.
    4 replies
    Gaurav Bhattacharjee
    @guilefoylegaurav

    It was only recently that I discovered wink nlp and I could not help but succumb to getting awestruck. I have also realized that the entire ecosystem uses no external dependencies, which has made it all the more interesting.

    I am extremely interested in contributing and have been studying the code. I was curious to know to know how the NER works, for a start. I have been reading the code for the gazetter based NER but would appreciate a walkthrough for clarity, as it would greatly aid to my comprehension of the code.

    Rachna Chakraborty
    @rachnachakraborty
    Hi @guilefoylegaurav delighted to know that you liked wink-nlp and would like to contribute. It would be a good idea to begin with reading the wink-nlp's user documentation. You can read more about contribution here.
    Sean Brunnock
    @brunnock
    On https://winkjs.org/wink-nlp/item.html , there's a sentence on the bottom that states, "Some of the properties are language dependent such as it.stem." Is there an object called "it" or is it a typo?
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    @brunnock It is a typo. "Its" is a helper, here is an example:
    const winkNLP = require( 'wink-nlp' );
    const its = require( 'wink-nlp/src/its.js' );
    const model = require( 'wink-eng-lite-web-model' );
    const nlp = winkNLP( model );
    
    const text = 'My mother had a beautiful singing voice.';
    const doc = nlp.readDoc( text );
    // Print tokens.
    console.log( doc.tokens().out() );
    // Print each token's stem.
    console.log( doc.tokens().out( its.stem ) );

    The above code will print:

    ["My", "mother", "had", "a", "beautiful", "singing", "voice", "."]
    ["my", "mother", "had", "a", "beauti", "sing", "voic", "."]

    @brunnock

    4 replies
    soulglider009
    @soulglider009
    Does winkjs have multilingual lemmatization support, or only english?
    Rachna Chakraborty
    @rachnachakraborty
    Hi @soulglider009 Multi-lingual lemmatization is in our roadmap, as of now its available for English model only.
    Miguel Costa
    @jmig_costa_twitter
    Hi all. where can I find documentation on the default entities types? I need to understand the [DURATION] entity.
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    Hello @jmig_costa_twitter. Entities detected by default are listed at https://github.com/winkjs/wink-eng-lite-web-model#named-entity-recognition-ner
    Miguel Costa
    @jmig_costa_twitter
    image.png
    thank you @sanjayaksaxena. But where can I see how the [DURATION] works? ( the pattern or the regex )
    I'm asking this because I'm having some issues with it. I.E: between 2000 and 3000 dollars
    outputs { value: 'between 2000', type: 'DURATION' },
    I have 2 issues with the output:
    1 - between is a range not necessarily a duration
    2 - if we consider it a duration, then it should grab the second value: between X1 and X2
    Miguel Costa
    @jmig_costa_twitter
    this one is also interesting
    image.png
    Sanjaya Kumar Saxena
    @sanjayaksaxena
    @jmig_costa_twitter thanks for pointing out... looks like a training issue and will require investigation of the model training at our end... sometimes fixing the model may not be easy! however we will surely lookup and revert back to you... it may take sometime at our end.
    Miguel Costa
    @jmig_costa_twitter

    @sanjayaksaxena it is me that has to thank you for or amazing work on winkjs. Let me know if I can help with anything. I have sent you a linkedin connection request, lets keep in touch.

    Once again thank you (and the team) for winkjs. You guys are doing an amazing work!

    Rachna Chakraborty
    @rachnachakraborty
    @jmig_costa_twitter Thanks for the kind and inspiring words.