Skip to content

Latest commit

 

History

History
204 lines (183 loc) · 5.62 KB

ner-manager.md

File metadata and controls

204 lines (183 loc) · 5.62 KB

NER Manager

The Named Entity Recognition manager is able to store an structure of entities and options of the entity for each language. Then, given an utterance and the language, is able to search the options of the entity inside the utterance, and return a list of the bests substrings. This is done using a threshold for the accuracy, by default the accuracy is 0.8 but you can provide it in the options when creating the instance.

Enum Named Entities

It support Enum Named Entities: Named entities where you define options, and the possible text for each option.

const { NerManager } = require('node-nlp');

const manager = new NerManager({ threshold: 0.8 });
manager.addNamedEntityText(
  'hero',
  'spiderman',
  ['en'],
  ['Spiderman', 'Spider-man'],
);
manager.addNamedEntityText(
  'hero',
  'iron man',
  ['en'],
  ['iron man', 'iron-man'],
);
manager.addNamedEntityText('hero', 'thor', ['en'], ['Thor']);
manager.addNamedEntityText(
  'food',
  'burguer',
  ['en'],
  ['Burguer', 'Hamburguer'],
);
manager.addNamedEntityText('food', 'pizza', ['en'], ['pizza']);
manager.addNamedEntityText('food', 'pasta', ['en'], ['Pasta', 'spaghetti']);
manager.findEntities(
  'I saw spederman eating speghetti in the city',
  'en',
).then(entities => {
  // ...
})
// value is [ { start: 6, end: 15, levenshtein: 1, accuracy: 0.8888888888888888, option: 'spiderman',
//  sourceText: 'Spiderman', entity: 'hero', utteranceText: 'spederman' },
//  { start: 23, end: 32, levenshtein: 1, accuracy: 0.8888888888888888, option: 'pasta',
//  sourceText: 'spaghetti', entity: 'food', utteranceText: 'speghetti' } ]

Regular Expression Named Entities

It also support Regular Expression entities

const { NerManager } = require('node-nlp');

const manager = new NerManager({ threshold: 0.8 });
manager.addNamedEntityText(
  'hero',
  'spiderman',
  ['en'],
  ['Spiderman', 'Spider-man'],
);
manager.addNamedEntityText(
  'hero',
  'iron man',
  ['en'],
  ['iron man', 'iron-man'],
);
manager.addNamedEntityText('hero', 'thor', ['en'], ['Thor']);
manager.addNamedEntityText(
  'food',
  'burguer',
  ['en'],
  ['Burguer', 'Hamburguer'],
);
manager.addNamedEntityText('food', 'pizza', ['en'], ['pizza']);
manager.addNamedEntityText('food', 'pasta', ['en'], ['Pasta', 'spaghetti']);
const entity = manager.addNamedEntity('email', 'regex');
entity.addRegex('en', /\b(\w[-._\w]*\w@\w[-._\w]*\w\.\w{2,3})\b/gi);
manager.findEntities(
  'I saw spiderman eating speghetti in the city and his mail is [email protected]',
  'en',
).then(entities => console.log(entities));
// [ { start: 6,
//     end: 15,
//     levenshtein: 0,
//     accuracy: 1,
//     option: 'spiderman',
//     sourceText: 'Spiderman',
//     entity: 'hero',
//     utteranceText: 'spiderman' },
//   { start: 23,
//     end: 32,
//     levenshtein: 1,
//     accuracy: 0.8888888888888888,
//     option: 'pasta',
//     sourceText: 'spaghetti',
//     entity: 'food',
//     utteranceText: 'speghetti' },
//   { start: 61,
//     end: 80,
//     accuracy: 1,
//     sourceText: '[email protected]',
//     utteranceText: '[email protected]',
//     entity: 'email' } ]

Trim Named Entities

It supports entities that works trimming text conditions, like text between two words. It supports 7 different conditions:

  • Between
  • After
  • After First
  • After Last
  • Before
  • Before First
  • Before Last
const { NerManager } = require('node-nlp');

const manager = new NerManager({ threshold: 0.8 });
const fromEntity = manager.addNamedEntity('fromEntity', 'trim');
fromEntity.addBetweenCondition('en', 'from', 'to');
fromEntity.addAfterLastCondition('en', 'to');
const toEntity = manager.addNamedEntity('toEntity', 'trim');
fromEntity.addBetweenCondition('en', 'to', 'from');
fromEntity.addAfterLastCondition('en', 'from');
manager.findEntities(
  'I want to travel from Barcelona to Madrid',
  'en',
).then(entities => console.log(entities));
// [ { type: 'between',
//     start: 22,
//     end: 31,
//     accuracy: 1,
//     sourceText: 'Barcelona',
//     utteranceText: 'Barcelona',
//     entity: 'fromEntity' },
//   { type: 'afterLast',
//     start: 35,
//     end: 41,
//     accuracy: 0.99,
//     sourceText: 'Madrid',
//     utteranceText: 'Madrid',
//     entity: 'fromEntity' },
//   { type: 'between',
//     start: 10,
//     end: 16,
//     accuracy: 1,
//     sourceText: 'travel',
//     utteranceText: 'travel',
//     entity: 'fromEntity' } ]

Utterances with duplicated Entities

Utterances with more than one entity with the same name are supported, providing a "whitelist" with numbered entity names.

The "numbered entity" format must be in the form ${entityName}_${integer}. E.g. "hero_1", "food_2", etc.

const { NerManager } = require('node-nlp');

const manager = new NerManager();

manager.addNamedEntityText('hero', 'spiderman', ['en'], ['Spider-man']);
manager.addNamedEntityText('hero', 'iron man', ['en'], ['ironman']);
manager.addNamedEntityText('food', 'pizza', ['en'], ['pizza']);
manager.addNamedEntityText('food', 'pasta', ['en'], ['spaghetti']);

const entities = manager.findEntities(
  'I saw spiderman eating spaghetti and ironman eating pizza',
  'en',
  ['hero_1', 'hero_2', 'food_1', 'food_2'],
);

// entities = [
//   {
//     entity: "hero_1",
//     option: "spiderman",
//     utteranceText: "spiderman",
//     ...
//   },
//   {
//     entity: "food_1",
//     option: "pasta",
//     utteranceText: "spaghetti",
//     ...
//   },
//   {
//     entity: "hero_2",
//     option: "iron man",
//     utteranceText: "ironman",
//     ...
//   },
//   {
//     entity: "food_2",
//     option: "pizza",
//     utteranceText: "pizza",
//     ...
//   }
// ]