Finding Repeated Structure in Time Series: The Most Useful Primitive in Data Science

Eamonn Keogh

University of California Riverside

Resumen/Abstract

Prof. Eamonn Keogh is a Distinguished Professor at the Department of Computer Science and Engineering in the University of California Riverside. He has invented many of the most commonly used primitives and representations for time series data mining, including SAX, PAA, Time Series Shapelets, Time Series Discords, Time Series Motifs, Time Series Chains and the Matrix Profile. The majority of papers on time series published in SIGKDD/ICDM/SIGMOD/VLDB etc. exploit at least one of his ideas or definitions. He has won at least one best paper award in all the major conferences in his area, including SIGKDD, ICDM,SIGMOD, SDM etc. His work is heavily cited; he has an H-index of 118 and a citation count of ~76,000. His research is widely used in industry, and has been funded by Google, Microsoft, IBM, Sony, Oracle, Mitsubishi, Vodafone, NetAPP, Samsung, AspenTech, Applied Materials and Siemens.

Curriculum ponente

Time series data mining is the task of finding patterns, regularities, and outliers in massive datasets. Given the ubiquity of time series in medicine, science, and industry, time series data mining is of increasing importance. In this talk I shall argue that the simple primitive of time series motif discovery, the task of finding approximately repeated patterns with a dataset, is the most useful core operation in all of time series data mining. In particular, it can be used as a primitive to enable many other useful tasks, such as summarization, segmentation, classification, clustering and anomaly detection. I will argue my case with examples of motif discovery in datasets as diverse as penguin behavior, cardiology, and astronomy.