KDD2013

Abstract: Mobile connected devices, and smartphones in particular, are rapidly emerging as a dominant computing and sensing platform. This poses several unique opportunities for data collection and analysis, as well as new challenges. In this tutorial, we survey the state-of-the-art in terms of mining data from mobile devices across different application areas such as ads, healthcare, geo-social, public policy, etc. Our tutorial has three parts. In part one, we summarize data collection in terms of various sensing modalities. In part two, we present cross-cutting challenges such as real-time analysis, security, and we outline cross-cutting methods for mobile data mining such as network inference, streaming algorithms, etc. In the last part, we specifically overview emerging and fast-growing application areas, such as noted above. Concluding, we briefly highlight the opportunities for joint design of new data collection techniques and analysis methods, suggesting additional directions for future research. Speaker bio for presenter 1: Spiros Papadimitriou is mainly interested in data mining for graphs and streaming data, clustering, time series, large-scale data processing, and mobile applications. His interests span from the very small (embedded devices, and sensors; Arduino) to the very large (large-scale data processing and analysis; Hadoop). He has published more than forty papers on these topics in refereed conferences and journals. He received the best paper award in SDM 2008, has three invited journal publications in best paper issues, several book chapters and he has filed multiple patents. He has also been invited to give keynote talks on graph and social network analysis (WAAMD 2008, and ADN 2009) and tutorials on time series stream mining (University of Maine Summer School, 2008) and large-scale analytics (Carnegie Mellon University, 2012). In the past, he has also developed and released a number of Android applications (including live-view mobile OCR, and web service clients) that have 50,000 downloads. He is currently an assistant professor at Rutgers University (MSIS-RBS). Prior to that, he was a research scientist at Google, and a research staff member at IBM Research. He was a Siebel scholarship recipient in 2005. He obtained his MSc and PhD degrees from Carnegie Mellon University. Speaker bio for presenter 2: Tina Eliassi-Rad is an Associate Professor of Computer Science at Rutgers University. Before joining academia, she was a Member of Technical Staff and Principal Investigator at Lawrence Livermore National Laboratory. Tina earned her Ph.D. in Computer Sciences (with a minor in Mathematical Statistics) at the University of Wisconsin-Madison. Within data mining and machine learning, Tinas research has been applied to the World-Wide Web, text corpora, large-scale scientific simulation data, complex networks, and cyber situational awareness. She has published over 50 peer-reviewed papers (including a best paper runner-up award at ICDM09 and a best interdisciplinary paper award at CIKM12); and has given over 70 invited presentations. Tina is an action editor for the Data Mining and Knowledge Discovery Journal. In 2010, she received an Outstanding Mentor Award from the US DOE Office of Science and a Directorate Gold Award from Lawrence Livermore National Laboratory for work on cyber situational awareness. For more details, visit http://eliassi.org.

Abstract: Large amounts of data arise in a multitude of situations, ranging from bioinformatics to astronomy, manufacturing, and medical applications. For concreteness our tutorial focuses on data obtained in the context of the internet, such as user generated content (microblogs, e-mails, messages), behavioral data (locations, interactions, clicks, queries), and graphs. Due to its magnitude, much of the challenges are to extract structure and interpretable models without the need for additional labels, i.e. to design effective unsupervised techniques. We present design patterns for hierarchical nonparametric Bayesian models, efficient inference algorithms, and modeling tools to describe salient aspects of the data. Dr. Amr Ahmed is a Research Scientist at Google. He received his PhD from Carnegie Mellon University in 2011. His thesis Modeling Users and Content: Structured Probabilistic Representation and Scalable Online Inference Algorithms was awarded the prestigious ACM SIGKDD Doctoral Dissertation award in 2012. He spent a year as a Research Scientist at Yahoo! Research before joining Google. He authored over 40 papers on topics that are core to this tutorial (including a best-paper runner-up award at WSDM 2012) and co-presented 3 tutorials at web and machine learning conferences. Dr. Alex Smola received his PhD from the University of Technology in Berlin in 1998. Subsequently he was research group leader and professor at the Australian National University and Senior Principal Researcher at National ICT Australia. From 2008 until 2012 he was Principal Research Scientist at Yahoo. Since 2012 he is a visiting researcher at Google and since 2013 a full professor at the Machine Learning Department of Carnegie Mellon University. He has written over 180 papers (that won several best paper awards at ICML, WSDM and SIGIR) and authored or edited 5 books. His work covers a broad range of subjects from statistical learning theory, convex optimization, and functional analysis to practical algorithms for scalable data classification, regression, clustering, and topic models. His recent work focuses on distributed, very large scale latent variable models for user profiling and content recommendation.