social media, crime prediction, gis, and enhanced nlp text analysis anthony corso claremont graduate...

Download Social Media, Crime Prediction, GIS, and Enhanced NLP Text Analysis Anthony Corso Claremont Graduate University Center for Information Systems and Technology

If you can't read please download the document

Upload: norah-oliver

Post on 17-Jan-2018

223 views

Category:

Documents


1 download

DESCRIPTION

Data Science Workflow Preparation Analysis Reflection Dissemination

TRANSCRIPT

Social Media, Crime Prediction, GIS, and Enhanced NLP Text Analysis Anthony Corso Claremont Graduate University Center for Information Systems and Technology GIS Social Media Crime Prediction Number of Colleges Seattle Salt Lake San Diego Dallas New Madrid Chicago Boston New Orleans Data Science Workflow Preparation Analysis Reflection Dissemination Data Science Workflow Preparation Social Media Corpus Crime Data ArcGIS Solution SNAP Data Acquire Data SNAP Retailers Acquire Data City of Chicago Crime Acquire Data Social Media Data Twitter API Tweets(8) Twitter Streams~ (1) billion Tweets collected over (10) months stored both locally and on AWS Tweet Corpus Acquire Data Twitter Streaming Parameters Geographic Area / Bounding Box Twitter-Specific API Information Clean Data Tweet Preprocessing Raw Data Clean Data Tweet Preprocessing Unit of Measure Clean Data Tweet Preprocessing Text Analysis with GATE Clean Data Additional NLP Text Analysis with NLP Enhancements Clean Data Resultant Output NLP Geo-Tagged Tweets Data Science Workflow Analysis Analysis TweetsCrimeSNAP Retailers GIS Data Layers Analysis Just Points! Analysis NLP Geo-Tagged Tweets, Crime, and SNAP Analysis Hot Spot Analysis - NLP Geo-Tagged Tweets, Crime, and SNAP Data Science Workflow Reflection Explore Alternatives Iterate Next Steps Findings Data Science Workflow Dissemination Questions