Knowledge Discovery from Multi-Sourced Data

Chen Ye is currently an Associate Researcher at the School of Computer Science and Technology, Hangzhou Dianzi University, China. She received the Ph.D. degree in Computer Software and Theory from Harbin Institute of Technology, China. Her current research interests include data repairing, truth discovery, and crowdsourcing. She has won the ACM SIGMOD China Doctoral Dissertation Award in 2020. Hongzhi Wang is a Professor and Doctoral Supervisor at the School of Computer Science and Technology, Harbin Institute of Technology, China. His research interests include big data management and analysis, data quality, graph data management, and web data management. He has published more than 150 papers, and he is the Primary Investigator of more than 10 projects including three NSFC projects, and co-PI of 973, 863, and NSFC key projects. He was awarded as Microsoft fellowship, China Excellent Database Engineer, and IBM Ph.D. fellowship. Guojun Dai is now working in the School of Computer Science and Technology of Hangzhou Dianzi University, as the Head of the National Brain-Computer Collaborative Intelligent Technology International Joint Research Center, the director of the Institute of Computer Application Technology. His research interests include Internet of Things, industrial big data, network collaborative manufacturing, edge computing, brain-computer interface, cognitive computing, artificial intelligence. He has published over 50 research papers in top-quality international conferences and journals, particularly, INFOCOM, IEEE Transactions on Industrial Informatics, and IEEE Transactions on Mobile Computing.

Chapter 1 Introduction 1.1 Knowledge Discovery 1.2 Main Challenges 1.3 Book Overview Chapter 2 Functional-dependency-based truth discovery for isomorphic data 2.1 Handling independent constraints 2.2 Handling inter-related constraints 2.3 Inter-source data aggregation 2.4 Update source weights Chapter 3 Denial-constraint-based truth discovery for isomorphic data Describe the truth discovery strategies for isomorphic data based on denial constraints 4.1 Denial constraint transformation 4.2 Optimized solution 4.3 Scalable strategies Chapter 4 Pattern discovery for heterogeneous data 4.1 Problem definition for multi-source heterogeneous data 4.2 Optimization framework 4.3 PatternFinder algorithm 4.4 The optimized grouping strategy Chapter 5 Deep fact discovery for text data 5.1 Fact extraction via mining patterns 5.2 The CNN-LSTM architecture 5.3 The fact encoder and pattern embedding 5.4 Training and inference