School of
Information Technology and Electrical Engineering

Speaker: Mr Yang Li
Seminar Date: Thu, 17/10/2019 - 15:30
Venue: 78-631; Data Science MM Lab
Host: A/Prof Helen Huang; Dr Hongzhi Yin

Seminar Type:  PhD Confirmation Seminar

Abstract: 

Social media has been developing rapidly over last decade, which drives the advancement of the social data research on various applications, such as location-based recommendation, event detection and sentiment analysis. The massive amount of social data enables researchers to explore complex relationships among data items from social networks by developing advanced machine learning frameworks. However, it still remains challenging for conventional models to precisely capture latent relationships between items from complex social data due to data sparsity, data complexity and data heterogeneity. In this task, I will present my recent research in complex social data analytics for relationship discovery. Two deep learning models that are PoI Augmentation Sequence to Sequence (PA-Seq2Seq) Model and Interpretable Hierarchical Graph Neural Network (IH-GNN) will be presented, which are proposed for the applications of PoI Recommendation and Fashion Recommendation respectively.

Biography: 

Yang Li is a PhD candidate in the Data Science Group under the supervision of A/Prof Helen Huang and Dr Hongzhi Yin. He received his Master Degree in Computer Science from the University of Queensland in 2018. His research interests include graph neural network, deep learning and recommender systems.