Prof Xue Li: Social Media Analysis for Detection of Cyberbullying and Suicide Ideation

Join our seminar online at our Zoom meeting. 

About the Speaker

Prof Xue Li is a Professor in the School of Information Technology at The University of Queensland. Prof Xue Li’s work on Big Data Analytics has attracted wide attention. He was listed as one of 50 “Most Powerful People in Australia" on Big Data by the Financial Review - the Power Issue in 2015; received the Microsoft Start-Up Q Award in 2014, and won the Best use of Open Data – QLD Premier’s Awards for Open Data in 2014. His research interests are in data analytics, machine learning, and intelligent information systems.

Over his research career, Prof Li has been awarded with 10 ARC DP/LP/LIEF grants as a Chief Investigator (CI) and a few UQ VC Strategic Research grants. He is an Associate Editor of the Journal of Advanced Internet of Things. He also was an Editor on the Board of International Journal of Management of Information Systems & Technology. He has over 200 publications as monograph, edited books, book chapters, and journal and conference papers. His Google h-Index is 51, with more than 9,000 citations. He has successfully supervised 30 PhD students to graduation as their principal supervisor.

Abstract

This seminar discusses some of the approaches that we have developed for the detection of cyberbullying and suicide ideation. People share their information, feelings, and experiences with or without using their true identities on social media. We have experienced many problems such as spamming, fake news, cyberbullying, and misuse of social media for causing harm to individuals.

Suicide is a critical issue in modern society. Early detection of suicide ideation may help in preventing suicide attempts to save lives. Current suicidal ideation detection methods are mostly clinical methods based on the interaction between social workers or experts and the identified individuals. The research with machine learning techniques with feature engineering or deep learning for automatic detection based on online social contents, is still in its early stages.