Ken D. Gorro, Mary Jane G. Sabellano, Christian V. Maderazo, Angie M. Ceniza, Kim Gorro
Criminal activity discourages domestic and foreign direct investment even tourists that would like to explore the wonders of nature in the country. It reduces competitiveness and led to the reallocation of resources that will create uncertainty and inefficiency. With the increasing popularity of social media such as Facebook, sharing of information, particularly regarding crimes with other people would be easy. For this reason, this study aims to discover the commonly shared crime experiences and warning on Facebook in the Philippines. In this study, customized web scraper was created to harvest Facebook posts with crime experiences. An SVM Model was created for automatically identifying the type of crime experiences being shared on Facebook. A total of 1,600 Facebook posts were used as training set for the SVM model. The precision of the model is 89% and the recall is 90%.The result shows that Facebook has significant amount of Facebook post with crime experiences. © 2017 Association for Computing Machinery.
University of San Carlos, School of Arts and Science, University of San Carlos, School of Arts and Science, Department of Computer and Information Sciences, Philippines
