Angus Hugh Dempster, Monash University
Time series data is ubiquitous, and includes data from medical, environmental, and industrial monitoring, motion sensor data, and data from financial markets. When we have a large quantity of time series data, a critical task is to learn to assign time series to different categories or classes. For example, we might want to recognise gestures or activities based on motion sensor data, or to identify land use based on periodic satellite imagery. However, many of the most accurate methods for time series classification are also the most computationally expensive. This talk will present several fundamentally faster and more scalable approaches to time series classification. These methods allow us to learn from larger quantities of data with much lower computational cost.
Biography
Angus is a research fellow at Monash University, where he also received his PhD. His research interests include scalable methods for time series classification, benchmarking and performance evaluation, and fundamental methods for machine learning.