Spelling Correction for Improved Detection of Malicious Chat Messages

Cyberbullying, sexting, profanity and other forms of malicious chat messages have become increasingly common in online virtual worlds and social networks that are used by children and teenagers. These conversations are dangerous to children. The partner organization has already implemented a rule based filtering system to filter out malicious messages. However, not all the malicious messages can be filtered out since people invent subtler forms of malicious messages in an effort to subvert such filtering systems. This project will employ the spelling correction techniques to convert the subtler forms of malicious messages into their original forms. It can improve the performance of filtering system to be more robust against subtler forms of malicious messages.

Faculty Supervisor:

Anoop Sarkar

Student:

Zhelun Wu

Partner:

Two Hat Security Research Corp.

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Simon Fraser University