{"id":5361,"date":"2017-05-11T23:28:43","date_gmt":"2017-05-11T23:28:43","guid":{"rendered":"https:\/\/www.rsoftsolution.com\/blog\/salesforce-aims-to-save-you-time-by-summarizing-emails-and-docs-with-machine-intelligence.html"},"modified":"2017-05-11T23:28:43","modified_gmt":"2017-05-11T23:28:43","slug":"salesforce-aims-to-save-you-time-by-summarizing-emails-and-docs-with-machine-intelligence","status":"publish","type":"post","link":"https:\/\/www.rbsitsoftwaresolution.in\/blog\/salesforce-aims-to-save-you-time-by-summarizing-emails-and-docs-with-machine-intelligence.html","title":{"rendered":"Salesforce aims to save you time by summarizing emails and docs with machine intelligence"},"content":{"rendered":"


\n<\/p>\n

We have all seen the studies \u2014 some American workers spend upwards of six hours a day handling email. It\u2019s not a great use of time,\u00a0it destroys productivity and it ultimately costs businesses money. A\u00a0new paper written by a team\u00a0Salesforce MetaMind researchers\u00a0<\/a>could eventually provide summaries of professional\u00a0communication. More effective text summarization tools would unlock serious\u00a0value for Salesforce\u00a0users \u2014\u00a0if the research community can finish working out the kinks.<\/p>\n

Using machine learning to produce text summaries is not easy, particularly when you\u2019re dealing with very long blocks of texts.\u00a0Methods that simply draw on the language of the source text to produce summaries are not very flexible and methods that generate completely new language often generate incoherent sentences.<\/p>\n

Salesforce attempts to improve the accuracy of doing\u00a0the later, \u00a0generating summaries with fresh language. The team\u2019s modifications to standard practice include the addition of reinforcement learning\u00a0and methods for reducing repetitive language and increasing the amount of context available to maximize accuracy.<\/p>\n

\"\"<\/p>\n

An example summary generated by Salesforce<\/p>\n

With reinforcement learning, \u00a0an optimal behavior is established \u2014 in this case maximizing accuracy as measured by a formalized test. The model is then asked to return successive summaries and each time the model receives an accuracy score, it adapts in an effort to receive a higher score the next time.<\/p>\n

A simple way to think about this is to imagine a situation where you had the opportunity to take a practice\u00a0exam in college with unlimited retakes. Each time you take the practice exam you modify your study strategy with the hope that you will maximize your outcome on the real exam. A human probably would only need a few attempts to get it right, but a machine needs considerably more for\u00a0trial and error.<\/p>\n

Reinforcement learning is\u00a0gradually becoming more common for tasks requiring language\u00a0generation. Beyond reinforcement, the modified\u00a0model also uses contextual information from the source document to aid in the generation of relevant new language and to reduce duplicated phrasing.<\/p>\n

\"\"Salesforce tried out its approach on the ROUGE test, short for Recall-Oriented Understudy for Gisting Evaluation. ROUGE is a collection of tests that enable fast analysis of the\u00a0accuracy of a generated summary.<\/p>\n

The tests compare snippets of generated summaries with snippets from accepted summaries. Variations of the test just attempt to match snippets of different lengths. Salesforce outperformed previous attempts with\u00a0two to three point gains. This might not seem like much, but in the world of machine learning that\u2019s fairly significant.<\/p>\n

As with all research, it\u2019s not quite ready for prime time yet. But the work is indicative of a few\u00a0things. In case it wasn\u2019t already obvious, Salesforce is serious about applying machine intelligence to the CRM.\u00a0And one of the company\u2019s early priorities is text summarization to support sales.<\/p>\n

<\/p>\n

Featured Image: Queensbury\/iStock\/Getty Images<\/small><\/p>\n

Article source: http:\/\/feedproxy.google.com\/~r\/techcrunchIt\/~3\/LnE4FFbESdA\/<\/p>","protected":false},"excerpt":{"rendered":"

We have all seen the studies \u2014 some American workers spend upwards of six hours a day handling email. It\u2019s not a great use of time,\u00a0it destroys productivity and it <\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[19],"tags":[25],"class_list":["post-5361","post","type-post","status-publish","format-standard","hentry","category-enterprise-software","tag-enterprise"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t