Social networks depict how users connect and interact with one another, and enable crucial predictive tasks, including node classification (NC), link prediction (LP), and community detection (CD). With the blooming and advances of deep learning, novel Graph Representation Learning (GRL) models, which learn the representations of nodes, are invented and widely applied on social networks. How can GRL benefit data science? In this talk, I will utilize our recent research outcomes to exhibit what, where, and how GRL can benefit a variety of tasks in data science. First, we will demonstrate that semantics-preserving GRL are able to significantly boost the performance of typical NC, LP, and CD tasks. Second, we will show that GRL can be applied to better model and exploit diverse relationships between various types of nodes in the realms of recommender systems and knowledge base. Third, through the tasks of fake news detection, air quality forecasting, and stock price prediction, we will further exhibit that GRL is still powerful even when the graphs cannot be observed.
Dr. Cheng-Te Li is now an Associate Professor at Institute of Data Science and Department of Statistics, National Cheng Kung University (NCKU), Tainan, Taiwan. He received his Ph.D. degree (2013) from Graduate Institute of Networking and Multimedia, National Taiwan University. Before joining NCKU, he was an Assistant Research Fellow (2014-2016) at CITI, Academia Sinica. Dr. Li’s research targets at Machine Learning, Data Mining, Social Networks and Social Media Analysis, Recommender Systems, and Natural Language Processing. He has a number of papers published at top conferences, including KDD, WWW, ICDM, CIKM, SIGIR, IJCAI, ACL, EMNLP, and ACM-MM. Dr. Li’s academic recognitions include 2020 MOST FutureTech Award, 2019 K. T. Li Young Researcher Award, 2018 MOST Young Scholar Fellowship (The Columbus Program), and 2016 Exploration Research Award of Pan Wen Yuan Foundation. He leads Networked Artificial Intelligence Laboratory (NetAI Lab) at NCKU.