Title: "Latent Variable Probabilistic Models for Text and Social Network Data"
ABSTRACT
Exploring and understanding large text and social network data sets is of increasing interest across multiple fields, in computer science, social science, history, and more. In this talk I will discuss recent work on probabilistic latent variable models for such data. Latent variable models have a long tradition in data analysis - they typically hypothesize the existence of simple unobserved phenomena to explain relatively complex observed data. The main part of the talk will review recent work on one such class of latent variable models for text data, namely statistical topic models (or latent Dirichlet allocation), including recent developments in applying such models to multiclass document classification and fast "online" algorithms for learning topics. We will also discuss briefly how similar types of latent variable models can be used to model social network data. A number of different data sets will be used during the talk to provide illustrative examples.
BIOGRAPHICAL NOTE
Padhraic Smyth is a Professor in the Department of Computer Science with a joint appointment in Statistics, as well as Director of the Center for Machine Learning and Intelligent Systems, all at the University of California, Irvine. His research interests include machine learning, data mining, pattern recognition, and applied statistics. He is an ACM Fellow (2013), a AAAI Fellow (2010), and a recipient of the ACM SIGKDD Innovation Award (2009). He is co-author of the text Modeling the Internet and the Web: Probabilistic Methods and Algorithms (with Pierre Baldi and Paolo Frasconi in 2003) and Principles of Data Mining, MIT Press (with David Hand and Heikki Mannila in 2001), and he served as program chair of the ACM SIGKDD 2011 conference and the UAI 2013 conference.
Padhraic has served in editorial and advisory positions for journals such as the Journal of Machine Learning Research, the Journal of the American Statistical Association, and the IEEE Transactions on Knowledge and Data Engineering. While at UC Irvine he has received research funding from agencies such as NSF, NIH, IARPA, NASA, and DOE, and from companies such as Google, IBM, Yahoo!, Experian, and Microsoft. In addition to his academic research he is also active in industry consulting and has worked with companies such as Samsung, eBay, Yahoo!, Microsoft, Oracle, Nokia, and AT&T, as well as serving as scientific advisor to local startups in Orange County. He also served as an academic advisor to Netflix for the Netflix prize competition from 2006 to 2009.
Padhraic received a first class honors degree in Electronic Engineering from National University of Ireland (Galway) in 1984, and the MSEE and PhD degrees (in 1985 and 1988 respectively) in Electrical Engineering from the California Institute of Technology. From 1988 to 1996 he was a Technical Group Leader at the Jet Propulsion Laboratory, Pasadena, and has been on the faculty at UC Irvine since 1996.
ABSTRACT
Exploring and understanding large text and social network data sets is of increasing interest across multiple fields, in computer science, social science, history, and more. In this talk I will discuss recent work on probabilistic latent variable models for such data. Latent variable models have a long tradition in data analysis - they typically hypothesize the existence of simple unobserved phenomena to explain relatively complex observed data. The main part of the talk will review recent work on one such class of latent variable models for text data, namely statistical topic models (or latent Dirichlet allocation), including recent developments in applying such models to multiclass document classification and fast "online" algorithms for learning topics. We will also discuss briefly how similar types of latent variable models can be used to model social network data. A number of different data sets will be used during the talk to provide illustrative examples.
BIOGRAPHICAL NOTE
Padhraic Smyth is a Professor in the Department of Computer Science with a joint appointment in Statistics, as well as Director of the Center for Machine Learning and Intelligent Systems, all at the University of California, Irvine. His research interests include machine learning, data mining, pattern recognition, and applied statistics. He is an ACM Fellow (2013), a AAAI Fellow (2010), and a recipient of the ACM SIGKDD Innovation Award (2009). He is co-author of the text Modeling the Internet and the Web: Probabilistic Methods and Algorithms (with Pierre Baldi and Paolo Frasconi in 2003) and Principles of Data Mining, MIT Press (with David Hand and Heikki Mannila in 2001), and he served as program chair of the ACM SIGKDD 2011 conference and the UAI 2013 conference.
Padhraic has served in editorial and advisory positions for journals such as the Journal of Machine Learning Research, the Journal of the American Statistical Association, and the IEEE Transactions on Knowledge and Data Engineering. While at UC Irvine he has received research funding from agencies such as NSF, NIH, IARPA, NASA, and DOE, and from companies such as Google, IBM, Yahoo!, Experian, and Microsoft. In addition to his academic research he is also active in industry consulting and has worked with companies such as Samsung, eBay, Yahoo!, Microsoft, Oracle, Nokia, and AT&T, as well as serving as scientific advisor to local startups in Orange County. He also served as an academic advisor to Netflix for the Netflix prize competition from 2006 to 2009.
Padhraic received a first class honors degree in Electronic Engineering from National University of Ireland (Galway) in 1984, and the MSEE and PhD degrees (in 1985 and 1988 respectively) in Electrical Engineering from the California Institute of Technology. From 1988 to 1996 he was a Technical Group Leader at the Jet Propulsion Laboratory, Pasadena, and has been on the faculty at UC Irvine since 1996. 
mobile