Tutorial Speaker |
Duong Tuan AnhShort BiographyDuong Tuan Anh holds Doctorate of Engineering in Computer Science from School of Advanced Technologies at the Asian Institute of Technology in Bangkok, Thailand where he also received his Master of Engineering in the same branch. He is currently Associate Professor at Faculty of Computer Science and Engineering, Ho Chi Minh City University of Technology. His research focuses on temporal databases, constraint programming, metaheuristics, and time series data mining. He is currently the head of Time Series Data Mining Research Group in his faculty. He authored more than 60 scientific papers. |
Tutorial Time and Date |
9h30-11h30 a.m, Wed 26th December, 2012 |
Turorial Venue |
Room 201B9, Ho Chi Minh City University of Technology, 268 Ly Thuong Kiet, District 10, Ho Chi Minh City. |
Title |
Discovering Motifs in Time Series Data |
Abstract |
A time series is a sequence of real numbers measured at equal intervals. Time series data arise in so many applications of various areas ranging from science, engineering, business, finance, economic, medicine to government. Besides similarity search, there are several other important time series data mining tasks such as classification, clustering, anomaly detection, finding motifs rule discovery and visualization. Time series motifs are frequently occurring but unknown subsequences in a longer time series. Discovering time series motifs is a crucial task in time series data mining. This task has been used to solve problems in various application areas and also used as a preprocessing step in several higher level data mining tasks. In this tutorial, we introduce some fundamental concepts on time series and similarity search in time series. We also review some state-of-the-art time series motif discovery algorithms and compare their effectiveness. Furthermore, we present our proposed method for time series motif discovery which is based on significant extreme points and clustering. This novel algorithm is much more effective than the widely used Random Projection algorithm in terms of time efficiency and accuracy. |
Presentation of the tutorial can be downloaded at here. |