wink-ner - npm Package Compare versions

Comparing version 0.1.0 to 0.2.0

package.json

 {
   "name": "wink-ner",
-  "version": "0.1.0",
-  "description": "Named Entity Recognition",
+  "version": "0.2.0",
+  "description": "Language agnostic named entity recognizer",
   "keywords": [
@@ -6,0 +6,0 @@ "NER",

src/wink-ner.js

@@ -228,2 +228,42 @@ //     wink-ner
 
+  // ### addUniWordEntity
+  /**
+   *
+   * Process uni-word entity and adds it to `uniWordEntities`.
+   *
+   * @param {string[]} words — from entity's text.
+   * @param {object} entity — to be added.
+   * @return {undefined} nothing!
+   * @private
+  */
+  var addUniWordEntity = function ( words, entity ) {
+    const firstWord = words[ 0 ];
+    let wordCounts;
+    // Latest value overrides previous value, if any.
+    if ( uniWordEntities[ firstWord ] ) wordCounts = uniWordEntities[ firstWord ].wordCounts;
+    uniWordEntities[ firstWord ] = cloneEntity( entity );
+    if ( wordCounts ) uniWordEntities[ firstWord ].wordCounts = wordCounts;
+  }; // addUniWordEntity()
+
+  // ### addUniWordEntity
+  /**
+   *
+   * Process multi-word entity and adds it to `multiWordEntities`.
+   *
+   * @param {string} text — property of entity.
+   * @param {string[]} words — from entity's text.
+   * @param {object} entity — to be added.
+   * @return {undefined} nothing!
+   * @private
+  */
+  var addMultiWordEntity = function ( text, words, entity ) {
+    const firstWord = words[ 0 ];
+    uniWordEntities[ firstWord ] = uniWordEntities[ firstWord ] || Object.create( null );
+    uniWordEntities[ firstWord ].wordCounts = uniWordEntities[ firstWord ].wordCounts || [];
+    if ( uniWordEntities[ firstWord ].wordCounts.indexOf( words.length ) === -1 ) uniWordEntities[ firstWord ].wordCounts.push( words.length );
+    multiWordEntities[ words.join( ' ' ) ] = cloneEntity( entity );
+    // The expression is a simple arithmatic formulation to detect acronyms.
+    if ( words.length === ( ( text.length + 1 ) / 2 ) ) addUniWordEntity( [ words.join( '' ) ], entity );
+  }; // addMultiWordEntity()
+
   // ### learn
@@ -243,2 +283,6 @@ /**
    *
+   * Acronyms must be added with space between each character; for example USA
+   * should be added as `'u s a'` — this ensure correct detection of
+   * `U S A` or `U. S. A.` or `U.S.A.` as `USA.`
+   *
    * @param {object[]} entities — where each element defines an entity via
@@ -270,2 +314,9 @@ * two mandatory properties viz. `text` and `entityType` as described later.
    * @example
+   * var trainingData = [
+   *   { text: 'manchester united', entityType: 'club' },
+   *   { text: 'manchester', entityType: 'city' },
+   *   { text: 'uk', entityType: 'country' }
+   * ];
+   * learn( trainingData );
+   * // -> 3
   */
@@ -275,22 +326,16 @@ var learn = function ( entities ) {
     // declarations in the beginning.
+    let length = 0;
     for ( let i = 0, imax = entities.length; i < imax; i += 1 ) {
       const entity = entities[ i ];
-      const text = entity.text;
-      const entityType = entity.text;
-      let wordCounts;
+      // Normalize after removing extra white spaces; required for acronyms processing.
+      const text = normalize( ( entity.text ).trim().replace( /\s+/, ' ' ) );
+      const entityType = ( entity.text ).trim().replace( /\s+/, ' ' );
+      // Add if `text` and `entityType` are defined.
       if ( text && entityType ) {
-        const words = normalize( text.trim() ).split( /\s+/ );
-        const firstWord = words[ 0 ];
+        const words = text.split( /\s+/ );
+        length += 1;
         if ( words.length === 1 ) {
-          // Process uni-word entity.
-          // Latest value overrides previous value, if any.
-          if ( uniWordEntities[ firstWord ] ) wordCounts = uniWordEntities[ firstWord ].wordCounts;
-          uniWordEntities[ firstWord ] = cloneEntity( entity );
-          if ( wordCounts ) uniWordEntities[ firstWord ].wordCounts = wordCounts;
+          addUniWordEntity( words, entity );
         } else {
-          // Process multi-word entity.
-          uniWordEntities[ firstWord ] = uniWordEntities[ firstWord ] || Object.create( null );
-          uniWordEntities[ firstWord ].wordCounts = uniWordEntities[ firstWord ].wordCounts || [];
-          if ( uniWordEntities[ firstWord ].wordCounts.indexOf( words.length ) === -1 ) uniWordEntities[ firstWord ].wordCounts.push( words.length );
-          multiWordEntities[ words.join( ' ' ) ] = cloneEntity( entity );
+          addMultiWordEntity( text, words, entity );
         }
@@ -305,3 +350,3 @@ }
 
-    return entities.length;
+    return length;
   }; // learn()
@@ -308,0 +353,0 @@

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

57502

increased by2.87%

Lines of code

526

increased by8.9%

No dependency changes