Show simple item record

dc.contributor.advisorChan, Philip K.
dc.contributor.authorHu, Huizhong
dc.date.accessioned2016-01-08T16:50:46Z
dc.date.available2016-01-08T16:50:46Z
dc.date.issued2015-12
dc.identifier.urihttp://hdl.handle.net/11141/781
dc.descriptionThesis (M.S.) - Florida Institute of Technology, 2015en_US
dc.description.abstractWith the increasing number of smartphone penetration, mining smartphone data that make smartphone smarter became a top research area, there are a lot of event data which we can use to predict behavior or detect anomalies. The privacy disclosure caused by stolen or lost phones becomes an increasingly difficult problem that cannot be ignored. So we design an anomaly detection system by mining patterns to detect stolen phones. We use a pattern mining algorithm to abstract patterns from user past behavior, then construct a personalized model and use a scoring function and threshold setting strategy to detect stolen events. Moreover, we apply our system to a data set from MIT Real Mining Project. Experimental results show that our system can detect 87% stolen events with 0.009 false positive rate on the average.en_US
dc.language.isoen_USen_US
dc.rightsCC BY Creative Commons with Attributionen_US
dc.rights.urihttp://creativecommons.org/licenses/by/3.0/en_US
dc.titleUsing a Personalized Machine Learning Approach to Detect Stolen Phonesen_US
dc.typeThesisen_US
thesis.degree.nameMasters of Science in Computer Scienceen_US
thesis.degree.levelMastersen_US
thesis.degree.disciplineComputer Scienceen_US
thesis.degree.departmentComputer Sciences and Cybersecurityen_US
thesis.degree.grantorFlorida Institute of Technologyen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

CC BY Creative Commons with Attribution
Except where otherwise noted, this item's license is described as CC BY Creative Commons with Attribution