讲座题目：Graph Data Management and GraphMachine Learning: Synergies and Opportunities

讲座时间：2024年10月30日 14:00-15:30

讲座地点：信息楼453

腾讯会议：445-926-173

报告摘要：

Theubiquity of machine learning, particularly deep learning, applied to graphs isevident in applications ranging from cheminformatics (drug discovery) andbioinformatics (protein interaction prediction) to knowledge graph-based queryanswering, fraud detection, and social network analysis. Concurrently, graphdata management deals with the research and development of effective,efficient, scalable, robust, and user-friendly systems and algorithms forstoring, processing, and analyzing vast quantities of heterogeneous and complexgraph data. This talk provides a comprehensive overview of the synergiesbetween graph data management and graph machine learning, illustrating how theyintertwine and mutually reinforce each other across the entire spectrum of thegraph data science and machine learning pipeline. This talk highlights twocrucial aspects: (1) How graph data management enhances graph machine learning,including contributions such as improved graph neural network performancethrough graph data cleaning, scalable graph embedding, efficient graph-basedvector data management, robust graph neural networks, user-friendly explainabilitymethods; and (2) how graph machine learning, in turn, aids in graph datamanagement, with a focus on applications like query answering over knowledgegraphs and various data science tasks. We discuss pertinent open problems anddelineate crucial research directions.

主讲嘉宾：

柯翔宇，浙江大学计算机科学与技术学院、软件学院平台“百人计划”研究员（2022至今），博士生导师，博士毕业于新加坡南洋理工大学（2017-2020），并先后于新加坡南洋理工大学（2020-2021）和新加坡国立大学（2021-2022）进行博士后研究，长期从事大数据智能管理与分析领域的研究工作，任中国计算机学会（CCF）数据库专委执行委员。柯翔宇博士在数据库、数据挖掘领域顶级国际会议和期刊上发表论文27篇，获宁波市（副省级）甬江人才工程科技创新领域青年创新人才项目支持，任宁波市科技项目评审专家，宁波市拔尖人才；作为主要参与人参与国家自然科学基金面上项目一项，浙江省“尖兵领雁”重点研发项目一项，宁波市数字孪生重点研发专项一项。担任VLDBJ、TKDE、TKDD等国内外顶级学术期刊评审专家，KDD、WWW、IJCAI、ICDE等顶级国际学术会议程序委员会委员。