BEGIN:VCALENDAR PRODID;X-RICAL-TZSOURCE=TZINFO:-//Calagator//EN CALSCALE:GREGORIAN X-WR-CALNAME:Calagator METHOD:PUBLISH VERSION:2.0 BEGIN:VTIMEZONE TZID;X-RICAL-TZSOURCE=TZINFO:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20160313T020000 RDATE:20160313T020000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 TZNAME:PDT END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT CREATED;VALUE=DATE-TIME:20160225T195413Z DTEND;TZID=America/Los_Angeles;VALUE=DATE-TIME:20160405T200000 DTSTART;TZID=America/Los_Angeles;VALUE=DATE-TIME:20160405T183000 DTSTAMP;VALUE=DATE-TIME:20160225T195413Z LAST-MODIFIED;VALUE=DATE-TIME:20160225T195413Z UID:http://calagator.org/events/1250469878 DESCRIPTION:Speaker: Winston Saunders \;\n \;\nAbstract: Word vecto rs\, derived by deep learning algorithms applied to billions of words of text\, provide powerful semantic models of language. Code in R\, demons trating [queen] + [man] - [woman] ~ [King] to about 90% accuracy will be reviewed. Building first on exploratory "\;bag of words"\; anal ysis of Presidential debate texts\, we'll explore\, using pre-computed G loVe vectors (Pennington et al http://nlp.stanford.edu/projects/glove/)\ , relationships like [sanders] + [trump] - [clinton] ~ [cruz] and how ca ndidate positions align to rhetorical sentiment like [government] + [peo ple] - [tax]. This analysis is work in progress. We'll also test empiric al limits (aka failed experiments). Active feedback is both sought and w elcome. \;\n \;\nDetails: \;\n \;\nDoors open ~6 pm\, talk s tarts at 6:30 pm \;\n \;\nDoors are open at bottom\, take elevator to 3rd floor\, door should be open for suite 320 \;\n \;\nWe'll v isit a local watering hole afterwards. \n\nTags: text mining\, data scie nce\, programming\, r\, politics\n\nImported from: http://calagator.org/ events/1250469878 URL:http://pdxrlang.com/ SUMMARY:pdxrlang meetup: NLP meets Politics-Experiment­s w/ Word-Vectors and 2016 Campaign Debate Rhetoric LOCATION:Mozilla: 1120 NW Couch St.\, Suite 320\, Portland Oregon 97209 U S SEQUENCE:1 END:VEVENT END:VCALENDAR