Invited Seminar Talk
Prof. Abdullah Mueen, University of New Mexico (USA)
when: 12 December 2016, 4:45pm
where: room Conseil, ground floor, Paris Descartes University, 12 rue de l'Ecole de Medecine, Paris 75006
Time series patterns are waveforms with properties useful for various data mining tasks such as summarization, classification and anomaly detection. In this talk, I present three primitive temporal patterns: Motifs, Shapelets, and Discords. Motifs are repeating segments in seemingly random time series data; Shapelets are small segments of long time series characterizing their sources; Discords are anomalous waveforms in long time series that do not repeat anywhere else. I briefly discuss efficient algorithms to discover these patterns and present cases in mining data from robots, humans and social media. Applications include activity classification using accelerometer and brain activity data, correlated clusters in social media data, and anomaly in online review data.
Abdullah Mueen is an Assistant Professor in Computer Science at University of New Mexico since 2013. Previously he was a Scientist in the Cloud and Information Science Lab at Microsoft Corporation. His major interest is in temporal data mining with a focus on two unique types of signals: social media and electrical sensors. He has been actively publishing in the data mining conferences including KDD, ICDM and SDM and journals including DMKD and KAIS. He has received runner-up award in the Doctoral Dissertation Contest in KDD 2012. He has won the best paper award in the same conference. Earlier, he earned PhD degree at the University of California at Riverside and BSc degree at Bangladesh University of Engineering and Technology.
Hosted by: Themis Palpanas
List of past seminars