讲课题目：The Long Pattern Mining Problem and its Applications
讲座内容摘要： Data mining is instrumental in today’s data-intensive science tackling the challenges of data deluge. Frequent pattern mining is a central topic in data mining. Patterns could be subsets, subsequences or substructures. Frequent patterns are the ones occurring frequently in a dataset. Over the past two decades, frequent pattern mining has drawn unparalleled attention and numerous algorithms have been proposed. Virtually, all such algorithms strive to mine frequent patterns with respect to a minimum frequency threshold, and infrequent ones below the threshold are systematically pruned off for efficiency. But, would those discarded infrequent patterns ever be interesting and useful? In this talk, we discuss the statistical and practical significance of previously ignored long patterns, as well as principled solutions that lead to efficient mining of such patterns from massive data.
讲座人概况：Byron J. Gao received Ph.D. and B.Sc. in Computer Science from Simon Fraser University, Canada, in 2007 and 2003 respectively. He was a postdoctoral fellow at the University of Wisconsin before joining Texas State University in 2008. His research spans several related fields of data mining, databases, and information retrieval. He constantly publishes and serves on reputable international conferences and journals. His research has been supported by the National Science Foundation, Department of Energy, and Texas Higher Education Coordinating Board.