Show simple item record

dc.creatorRao, Akash R
dc.date.accessioned2021-07-24T00:33:26Z
dc.date.available2021-07-24T00:33:26Z
dc.date.created2021-05
dc.date.submittedMay 2021
dc.identifier.urihttps://hdl.handle.net/1969.1/194438
dc.description.abstractSocial media has grown to become a fundamental part of our lives over the past two decades and with its growth, the misuse of the platform for extremist purposes has become common. The wide reach of social media has allowed extremist groups to take advantage of the platform to spread terrorist propaganda and fear. Therefore, the need for a robust extremist detector in social media is evident. As an attempt to combat this problem, we present techniques to detect various forms of extremism in videos crawled from Twitter, a social media to share short posts. We build upon existing deep neural networks used for action classification and create a model capable of recog- nizing certain common extremism types. Additionally, we also expand on logo/object detection models for the same purpose. We then use these models against a sample space of roughly 2 million unlabelled videos to test the accuracy of these models.
dc.format.mimetypeapplication/pdf
dc.subjectextremism detection
dc.subjectterrorist propaganda
dc.subjectsocial media
dc.subjectlogo detection
dc.subjectaction classification
dc.titleExtremism Video Detection In Social Media
dc.typeThesis
thesis.degree.departmentComputer Science and Engineering
thesis.degree.disciplineComputer Science
thesis.degree.grantorUndergraduate Research Scholars Program
thesis.degree.nameB.S.
thesis.degree.levelUndergraduate
dc.contributor.committeeMemberCaverlee, James
dc.type.materialtext
dc.date.updated2021-07-24T00:33:27Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record