We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Describe the bug Named entity recognizer is failing on Turkish keywords even on 1/1 matching
To Reproduce
const { NerManager } = require('node-nlp'); const manager = new NerManager({ threshold: 0.5 }); manager.addNamedEntityText( 'entity', "Sağlık", ['tr'], ["ilaç"] ); manager.addNamedEntityText( 'currency', 'currency', ['tr'], ["türk lira", "lira", "₺", "tl", "tele", "dolar", "usd", "$", "euro", "avro", "yuro", "oyro", "€", "usd", "eur", "gbp", "jpy", "cnh", "aud", "dkk", "sek", "chf", "cad", "kwd", "nok", "sar"], ); const entity = manager.addNamedEntity('amount', 'regex'); entity.addRegex('tr', /\b([0-9]+(?:[\.\,][0-9]{1,3})?)\s*(t+[ÜüUu]*r+[KkQq]+\ +l+[Iİıi1]*r+[Aa@4]+|l+[Iİıi1]*r+[Aa@4]+(?:[^\s]*)|\₺+|t+l+|t+[Ee€]*l+[Ee€]+|d+[OÖoö]*l+[Aa@4]*r+|[ÜüUu]+[SsŞş]+d+|\$+|[Ee€]+[ÜüUu]*r+[OÖoö]+|[Aa@4]+[vVwW]+r+[OÖoö]+|y+[ÜüUu]*r+[OÖoö]+|[OÖoö]+y+r+[OÖoö]+|\€+|[ÜüUu]+[SsŞş]+d+|[Ee€]+[ÜüUu]*r+|[ĞğGg]+[Bbß]+p+|j+p+y+|[ÇçCc¢]+n+h+|[Aa@4]+[ÜüUu]*d+|d+[KkQq]+[KkQq]+|[SsŞş]+[Ee€]*[KkQq]+|[ÇçCc¢]+h+f+|[ÇçCc¢]+[Aa@4]*d+|[KkQq]+[vVwW]+d+|n+[OÖoö]*[KkQq]+|[SsŞş]+[Aa@4]*r+)\b/gi); manager.findEntities( '10 tl ilaç', 'tr', ).then(entities => { console.log(entities); });
Expected behavior it should find "ilaç" keyword with %100 matching but it's getting it with %75
Here's the result of the above lines:
[ { start: 6, end: 8, len: 3, levenshtein: 1, accuracy: 0.75, option: 'Sağlık', sourceText: 'ilaç', entity: 'entity', utteranceText: 'ila' }, { start: 3, end: 4, len: 2, levenshtein: 0, accuracy: 1, option: 'currency', sourceText: 'tl', entity: 'currency', utteranceText: 'tl' }, { start: 0, end: 5, accuracy: 1, sourceText: '10 tl', utteranceText: '10 tl', entity: 'amount' } ]
Desktop (please complete the following information):
The text was updated successfully, but these errors were encountered:
No branches or pull requests
Describe the bug
Named entity recognizer is failing on Turkish keywords even on 1/1 matching
To Reproduce
Expected behavior
it should find "ilaç" keyword with %100 matching but it's getting it with %75
Here's the result of the above lines:
Desktop (please complete the following information):
The text was updated successfully, but these errors were encountered: