Comparing version 2.0.2 to 2.1.0
{ | ||
"name": "wink-ner", | ||
"version": "2.0.2", | ||
"version": "2.1.0", | ||
"description": "Language agnostic named entity recognizer", | ||
@@ -33,8 +33,8 @@ "keywords": [ | ||
"coveralls": "^3.0.3", | ||
"docdash": "winkjs/docdash", | ||
"docdash": "github:winkjs/docdash", | ||
"docker": "^1.0.0", | ||
"eslint": "^5.16.0", | ||
"eslint": "^6.8.0", | ||
"istanbul": "^1.1.0-alpha.1", | ||
"jsdoc": "^3.5.5", | ||
"mocha": "^6.1.3", | ||
"mocha": "^7.0.0", | ||
"mocha-lcov-reporter": "^1.3.0" | ||
@@ -41,0 +41,0 @@ }, |
@@ -22,3 +22,3 @@ # wink-ner | ||
var ner = require( 'wink-ner' ); | ||
// Create your instance of wink ner & use defualt config. | ||
// Create your instance of wink ner & use default config. | ||
var myNER = ner(); | ||
@@ -100,4 +100,4 @@ // Define training data. | ||
**wink-ner** is copyright 2017-19 [GRAYPE Systems Private Limited](http://graype.in/). | ||
**wink-ner** is copyright 2017-20 [GRAYPE Systems Private Limited](http://graype.in/). | ||
It is licensed under the terms of the MIT License. |
// wink-ner | ||
// Language agnostic named entity recognizer | ||
// | ||
// Copyright (C) 2017-19 GRAYPE Systems Private Limited | ||
// Copyright (C) 2017-20 GRAYPE Systems Private Limited | ||
// | ||
@@ -141,2 +141,3 @@ // This file is part of “wink-ner”. | ||
validTags.unknown = true; | ||
validTags.symbol = true; | ||
@@ -408,40 +409,31 @@ cfg.tagsToIgnore = Object.create( null ); | ||
// If ignorable then simply follow the next iteration, after saving the | ||
// current value into the original sequence of tokens; otherwise if | ||
// token is a word or number save it in words else just reset & break! | ||
// current value into the original sequence of tokens; otherwise | ||
// save it in words else just reset & break! | ||
if ( isIgnorable( value, tag ) ) { | ||
originalSeq.push( value ); | ||
} else if ( tag === 'word' || tag === 'number' ) { | ||
// `word` and `number` tags form entities, include them in `words`, | ||
// while ensuring `words` receive normalized value. | ||
words.push( normalize( value ) ); | ||
// Here `value` must go **as is**. | ||
originalSeq.push( value ); | ||
} else { | ||
// Can not be ignored: reset & break! | ||
words = []; | ||
originalSeq = []; | ||
break; | ||
} | ||
} else { | ||
// Ensure `words` receive normalized value. | ||
words.push( normalize( value ) ); | ||
// Here `value` must go **as is**. | ||
originalSeq.push( value ); | ||
} | ||
} // for ( k = i... ) | ||
if ( words.length > 0 ) { | ||
entity = multiWordEntities[ words.join( ' ' ) ]; | ||
if ( !entity ) { | ||
words.push( singularize( words.pop() ) ); | ||
entity = multiWordEntities[ words.join( ' ' ) ]; | ||
if ( !entity ) { | ||
words.push( singularize( words.pop() ) ); | ||
} | ||
entity = multiWordEntities[ words.join( ' ' ) ]; | ||
} | ||
if ( entity ) { | ||
// Copy the entity into the merged token. | ||
mergedToken = copyKVPs( Object.create( null ), entity ); | ||
// Save original sequence of tokens that were detected as an entity. | ||
mergedToken.originalSeq = originalSeq; | ||
// If id is missing, create id by joining `words`. | ||
mergedToken.uid = entity.uid || words.join( '_' ); | ||
// If value was not defined, default to words. | ||
mergedToken.value = mergedToken.value || words.join( ' ' ); | ||
// move the next token's index — `nti` to `k`. | ||
return { token: mergedToken, nti: k }; | ||
} | ||
if ( entity ) { | ||
// Copy the entity into the merged token. | ||
mergedToken = copyKVPs( Object.create( null ), entity ); | ||
// Save original sequence of tokens that were detected as an entity. | ||
mergedToken.originalSeq = originalSeq; | ||
// If id is missing, create id by joining `words`. | ||
mergedToken.uid = entity.uid || words.join( '_' ); | ||
// If value was not defined, default to words. | ||
mergedToken.value = mergedToken.value || words.join( ' ' ); | ||
// move the next token's index — `nti` to `k`. | ||
return { token: mergedToken, nti: k }; | ||
} | ||
@@ -529,3 +521,3 @@ | ||
t = tokens[ i ]; | ||
if ( t.tag === 'word' ) { | ||
if ( !cfg.tagsToIgnore || !cfg.tagsToIgnore[ t.tag ] ) { | ||
// Look up for the word; if not found try its base form. | ||
@@ -720,1 +712,10 @@ let value = t.value; | ||
module.exports = ner; | ||
var x = ner(); | ||
var t = require( 'wink-tokenizer' )().tokenize; | ||
x.learn( | ||
[ | ||
{ text: 'f - 16', entityType: 'plane' } | ||
] | ||
); | ||
x.recognize( t( '-p' ) ); |
Sorry, the diff of this file is not supported yet
License Policy Violation
LicenseThis package is not allowed per your license policy. Review the package's license to ensure compliance.
Found 1 instance in 1 package
License Policy Violation
LicenseThis package is not allowed per your license policy. Review the package's license to ensure compliance.
Found 1 instance in 1 package
722
45691