{"id":7609,"date":"2016-05-11T06:26:16","date_gmt":"2016-05-11T06:26:16","guid":{"rendered":"http:\/\/www.kurzweilai.net\/?p=280245"},"modified":"2016-05-14T23:30:50","modified_gmt":"2016-05-14T23:30:50","slug":"google-open-sources-natural-language-understanding-tools","status":"publish","type":"post","link":"https:\/\/hoo.central12.com\/fugic\/2016\/05\/11\/google-open-sources-natural-language-understanding-tools\/","title":{"rendered":"Google open-sources natural language understanding tools"},"content":{"rendered":"<p>Google has just <a href=\"http:\/\/googleresearch.blogspot.com\/2016\/05\/announcing-syntaxnet-worlds-most.html\" >released<\/a> two powerful natural language understanding tools for free, open-source use by anyone. These tools allow machines to read and understand English text (such as text you type into a browser to do a Google search).<\/p>\n<p><em><a href=\"https:\/\/github.com\/tensorflow\/models\/tree\/master\/syntaxnet\" >SyntaxNet<\/a><\/em> is a \u201csyntactic parser\u201d &#8212; it allows machines to parse, or break down, sentences into their component parts of speech and identify the underlying meaning). And the <em>Parsey McParseface<\/em> program implements SyntaxNet in English (it learned from an annotated collection of old newswire stories called <a href=\"https:\/\/www.cis.upenn.edu\/~treebank\/\" >The Penn Treebank Project<\/a>).<\/p>\n<p>Here&#8217;s an example of how it parses and analyzes an English sentence :<img style=' display: block; margin-right: auto; margin-left: auto;'  class=\"aligncenter\" src=\"https:\/\/2.bp.blogspot.com\/-fqtmVS97tOs\/VzTEAI9BQ8I\/AAAAAAAAA_U\/xPj0Av64sGseS0rF4Z1BbhmS77J-HuEvwCLcB\/s1600\/image04.gif\" alt=\"\" width=\"600\" height=\"316\" \/>Using deep neural networks, SyntaxNet is implemented in Google&#8217;s TensorFlow (see <a href=\"http:\/\/www.kurzweilai.net\/google-open-sources-its-tensorflow-machine-learning-system\" >Google open-sources its TensorFlow machine learning system<\/a>).<\/p>\n<p><strong>So how well does it work?<\/strong><\/p>\n<p>On a standard benchmark consisting of randomly drawn English newswire sentences (&#8220;Penn Treebank&#8221;), Parsey McParseface recovers individual dependencies between words with over 94% accuracy, Google says. \u201dLinguists trained for this task agree in 96&#8211;97% of the cases. This suggests that we are approaching human performance\u2014but only on well-formed text.<\/p>\n<p>&#8220;Because Parsey McParseface is the\u00a0<a href=\"http:\/\/arxiv.org\/abs\/1603.06042\" >most accurate such model in the world<\/a>, we hope that it will be useful to developers and researchers interested in automatic extraction of information, translation, and other core applications of NLU,&#8221; says Google.<\/p>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Google has just released two powerful natural language understanding tools for free, open-source use by anyone. These tools allow machines to read and understand English text (such as text you type into a browser to do a Google search). SyntaxNet is a &ldquo;syntactic parser&rdquo; &mdash; it allows machines to parse, or break down, sentences into [&#8230;]<\/p>\n","protected":false},"author":13,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[46,43],"tags":[],"class_list":["post-7609","post","type-post","status-publish","format-standard","hentry","category-airobotics","category-news"],"_links":{"self":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts\/7609"}],"collection":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/users\/13"}],"replies":[{"embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/comments?post=7609"}],"version-history":[{"count":3,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts\/7609\/revisions"}],"predecessor-version":[{"id":7661,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts\/7609\/revisions\/7661"}],"wp:attachment":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/media?parent=7609"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/categories?post=7609"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/tags?post=7609"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}