March 24, 2023

About the speaker

Dr. Yifeng Gao is an Assistant Professor in the Department of Computer Science at the University of Rio Grande Valley. 

He received his Ph.D. degree at George Mason University in 2021. His research has appeared in premier conferences and journals in the data mining research field, including IEEE International Conference on Data Mining (ICDM), AAAI Conference on Artificial Intelligence, SIAM International Conference on Data Mining (SDM),  European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (PPKD), International Conference on Extending Database Technology (EDBT), International Conference on Data Mining and Knowledge Discovery (DMKD), and Knowledge and Information Systems (KAIS) . 

His research interests include time series data mining, deep learning, motif discovery, symbolic representation, visualization, and anomaly detection. Representing the Association for Computing Machinery (ACM), he served as Web Co-chair of the Special Interest Group on Knowledge Discovery and Data Mining (SIGKDD) 2022, Undergraduate Consortium Co-chair at SIGKDD 2023, Co-chair at AI for Time Series Analysis, International Joint Conferences on Artificial Intelligence Organization (AI4TS-IJCAI) 2022, and Co-chair at AI4TS-IJCAI 2023. He served as program committee (PC) member at prestigious data mining and machine learning venues including International Conference on Learning Representations (ICLR), KDD, PKDD, SDM, ICDM, and IJCAI.  

From Repeating Pattern Mining to Evolving Pattern  Mining, Detecting Meaningful Time Series Chain in  Massive Time Series

Time series data is one of the most commonly encountered data types, touching almost every aspect of human life, including medicine, finance, sciences, and industry. With the widespread use of sensors, large-scale time series have become ubiquitous in various applications. The task of finding potential meaningful patterns in time series has become one of the most essential tasks in time series data mining. While previous work mainly focused on finding the most frequently repeated pattern that does not change over time, recently more and more research has started to focus on finding evolving patterns. Many efficient and robust algorithms have been introduced in the last ten years. In this talk we will first introduce recent research in time series motif discovery, and then we will introduce recent advance in finding the time series chain, the time-evolving pattern. We will also demonstrate how to use them in real-world applications.