Keynotes will be held in the Houston Ballroom
Alexander Tuzhilin, Stern School of Business, New York University
Opportunities and Challenges Facing Recommender Systems: Where Can We Go from Here?
The field of Recommender Systems has experienced extensive growth over the last decade both in the academia and the industry and has established itself as a vibrant and mature research area in data mining and other disciplines. Given all this progress and accomplishments achieved by the field, it is a good time to ask the where-can-we-go-from-here question in order to work on the next generation of recommender systems that will allow us to overcome the limitations that the current generation of these systems is facing.
This talk will address this question and present some of the underexplored directions in the field of recommender systems that present promising research opportunities according to the speaker’s perspective. It will also address the challenges that researchers and practitioners may face pursuing some of these research directions.
Alexander Tuzhilin is a Professor of Information Systems, the NEC Faculty Fellow and the Chair of the Department of Information, Operations and Management Sciences at the Stern School of Business at NYU. He has received Ph.D. in Computer Science from the Courant Institute of Mathematical Sciences, NYU. His current research interests include data mining, recommender systems and personalization. Dr. Tuzhilin has published extensively on these and other topics and has served on the organizing and program committees of numerous conferences, including as a Program Co-Chair of the Third IEEE International Conference on Data Mining (ICDM), as a Conference Co-Chair of the Third ACM Conference on Recommender Systems (RecSys), and as the Chair of the Steering Committee of the ACM Conference on Recommender Systems. He has also served on the Editorial Boards of the IEEE Transactions on Knowledge and Data Engineering, the Data Mining and Knowledge Discovery Journal, the ACM Transactions on Management Information Systems, the INFORMS Journal on Computing (as an Area Editor), the Electronic Commerce Research Journal and the Journal of the Association of Information Systems. Results of Dr. Tuzhilin’s various academic and industrial activities have been featured in major media publications, including The New York Times, The Wall Street Journal, Business Week and The Financial Times.
Joydeep Ghosh, University of Texas, Austin
Predictive Healthcare Analytics under Privacy Constraints
The move to electronic health records is producing a wealth of information, which has the potential of providing unprecedented insights into the cause, prevention, treatment and management of illnesses. Analyses of such data also promises numerous opportunities for much more effective and efficient delivery of healthcare. However (valid) privacy concerns and restrictions prevent unfettered access to such data. In this talk I will first provide a perspective on the privacy vs. utility trade-off in the context of healthcare analytics. I will then outline two approaches that we have recently and successfully taken that provide privacy-aware predictive modeling with little degradation in model quality despite restrictions on what can be shared or analyzed. The first approach focuses on extracting predictive value from data that has been aggregated at various levels due to privacy concerns, while the second introduces a novel, non-parametric sampler that can generate "realistic but not real" data given a dataset that cannot be shared as is.
Joydeep Ghosh is currently the Schlumberger Centennial Chair Professor of Electrical and Computer Engineering at the University of Texas, Austin. He joined the UT-Austin faculty in 1988 after being educated at, (B. Tech '83) and The University of Southern California (Ph.D’88). He is the founder-director of IDEAL (Intelligent Data Exploration and Analysis Lab) and a Fellow of the IEEE. Dr. Ghosh has taught graduate courses on data mining and web analytics every year to both UT students and to industry, for over a decade. He was voted as "Best Professor" in the Software Engineering Executive Education Program at UT.
Dr. Ghosh's research interests lie primarily in data mining and web mining, predictive modeling / predictive analytics, machine learning approaches such as adaptive multi-learner systems, and their applications to a wide variety of complex real-world problems. He has published more than 300 refereed papers and 50 book chapters, and co-edited over 20 books. His research has been supported by the NSF, Yahoo!, Google, ONR, ARO, AFOSR, Intel, IBM, and several others. He has received 14 Best Paper Awards over the years, including the 2005 Best Research Paper Award across UT and the 1992 Darlington Award given by the IEEE Circuits and Systems Society for the overall Best Paper in the areas of CAS/CAD. Dr. Ghosh has been a plenary/keynote speaker on several occasions such as MICAI'12, KDIR'10, ISIT'08, ANNIE’06 and MCS 2002, and has widely lectured on intelligent analysis of large-scale data. He served as the Conference Co-Chair or Program Co-Chair for several top data mining oriented conferences, including SDM'13, SDM''12, KDD 2011, CIDM’07, ICPR'08 (Pattern Recognition Track) and SDM'06. He was the Conf. Co-Chair for Artificial Neural Networks in Engineering (ANNIE)'93 to '96 and '99 to '03 and the founding chair of the Data Mining Tech. Committee of the IEEE Computational Intelligence Society. He has also co-organized workshops on high dimensional clustering, Web Analytics, Web Mining and Parallel/ Distributed Knowledge Discovery.
Jianchang (JC) Mao, Microsoft
Large-scale Learning in Computational Advertising
Online Advertising is one of the fastest growing businesses on the Internet today. Search engines, web publishers, major ad networks, and ad exchanges are now serving billions of ad impressions per day and generating hundreds of terabytes of user events data every day. The rapid growth of online advertising has created enormous opportunities as well as technical challenges that involve Big Data. Computational Advertising attempts to mine the big data for making optimal ads serving decision in order to maximize a total utility function that captures publisher revenue, user experience and return on investment for advertisers. It has emerged as a new interdisciplinary field that involves information retrieval, machine learning, data mining, statistics, operations research, and micro-economics, to solve challenging problems that arise in online advertising.
In this talk, I will outline a number of major big data learning problems in various aspects of computational advertising, including user/query intent understanding, document/ad understanding, user targeting, ad selection, relevance modeling, user response prediction, keyword recommendation, forecasting, allocation, and marketplace optimization. Then, I will showcase our recent solutions to some of these problems, including query clustering for auction optimization and keyword recommendation. Query clustering for auction optimization is based on KL-divergence between two queries represented by their rank-score distributions under the Gaussian mixture assumption. We derived a variational EM algorithm for minimizing an upper bound of the total within-cluster KL-divergence. These clusters are then used for optimizing auction parameters, which yields significant improvements in marketplace KPIs. Keyword recommendation is formulated as a supervised multi-label random forest learning problem where labels (categories) are tens of millions of keywords and training data is automatically generated from click logs. Large-scale experiments conducted with 50 million webpages and 10 million keywords extracted from Bing logs showed significant gains in precision at 10 compared to previous ranking and NLP based techniques.
Jianchang (JC) Mao is Partner & Head of Advertising Relevance and Revenue Development in the Applications and Services Group at Microsoft, responsible for R&D of technologies and products that power Paid Search and Display Marketplaces. He joined Microsoft in April 2012. Previously, Mao was Vice President and Head of Advertising Sciences at Yahoo! Labs, overseeing the R&D of advertising technologies and products. He was also the Science/Engineering Director responsible for the development of back-end technologies for several Yahoo! social search products including Yahoo! Answers. Prior to joining Yahoo!, Mao was Director of Emerging Technologies and Principal Architect at Verity Inc., a leader in enterprise search, from 2000 to 2004. Prior to this, he was a research staff member at the IBM Almaden Research Center from 1994 to 2000, after receiving his PhD degree in computer science from Michigan State University in 1994. At Yahoo!, Mao was a Master Inventor awarded in 2012, received the Leadership Superstar Award (for VP and above) in 2010, and received a Superstar Team Award in 2008. During his tenure at IBM Almaden Research Center, he received an IBM Outstanding Technical Achievement Award and several Research Division Awards for outstanding contributions.
Mao’s research interests include machine learning, data mining, information retrieval, computational advertising, social networks, pattern recognition, and image processing. He has published more than 50 papers in journals, book chapters, and conferences, and holds 25 U.S. patents. Mao received an Honorable Mention Award in ACM KDD Cup 2002 (Task 1: Information Extraction from Biomedical Articles), an IEEE Transactions on Neural Networks Outstanding Paper Award in 1996 (for his 1995 paper), and an Honorable Mention Award from the International Pattern Recognition Society in 1993. He served as an associate editor of the IEEE Transactions on Neural Networks (1999-2000). Mao received the Distinguished Alumni Award from the Computer Science and Engineering Department at Michigan State University in 2011. Mao is an IEEE Fellow.