Jia Liu, Ph. D Candidate in Marketing, CBS. I am a Ph. D. candidate in Marketing at Columbia Business School. I am mainly interested in building empirical models to understand and predict consumer behavior, with the objective of generating managerial insights and applications for marketersfirms. I am on the Marketing job market this year, and will attend the 2. Summer AMA Conference in Atlanta. My Ph. D dissertation research, advised by Olivier Toubia, is titled as Modeling Text Based Consumer Online Search Behavior Linking Search Queries with Content Preferences. This research is motivated by the basic question of how to learn consumers content preferences from their online search queries. There are three essays in my dissertation that address each of the following research questions. How do users formulate online search queries How to infer consumer content preferences from search queries How to learn and model the dynamics of both query formation and subsequent clicking behavior I approach these questions using a combination of experiments, large scale field data, and sophisticated statistical and machine learning models. Currently, the first two essays are invited for revisions at Marketing Science, and the thrid essays is in progress. Research Interests Substantive Consumer Online Search, Search Engine MarketingOptimization, Loyalty Programs, Consumer Bounded Rationality, Behavioral Economics, Learning. Methodological Topic Modeling, Natural Language Processing, Machine Learning, Variational Bayesian Methods, Experiments. Job Market Paper. My job market paper is based on the second essay of my doctoral dissertation which investigates the linkage between consumer content preferences and online search queries. The key challenges in modeling search query are the curse of dimensionality i. We address these issues, by developing an innovative topic model, Hierarchically Dual Latent Dirichlet Allocation HDLDA, to estimate consumers content preferences from search queries. To validate this model, we built our own customized search engine to collect search data in a controlled experiment in which users actual content preferences can be observed. Curse Of Dimensionality Pdf' title='Curse Of Dimensionality Pdf' />We find that our proposed approach outperforms benchmark models in predicting users content preferences. Liu, Jia, and Olivier Toubia. A Semantic Approach for Estimating Consumer Content Preference from Online Search Queries. Invited for revision at Marketing Science. PDF Previous version on SSRN. Working Papers Liu, Jia, and Olivier Toubia. How do Consumers Form Online Search Queries The Importance of Semantic Relationships Between Queries and Results. Invited for revision at Marketing Science. PDF Previous version on SSRN. Liu, Jia, Asim Ansari, and Leonard Lee. The Joint Effect of Revenue based Loyalty Programs and Promotions on Consumer Purchase Behavior. Preparing for submission at Journal of Marketing Research. Curse Of Dimensionality Pdf' title='Curse Of Dimensionality Pdf' />The main linear technique for dimensionality reduction, principal component analysis, performs a linear mapping of the data to a lowerdimensional space in such a way. Confusion Matrixbased Feature Selection Soa Visa Computer Science Department College of Wooster Wooster, OH svisawooster. Brian Ramsay Sentry Data Systems, Inc. Publishes papers reporting new theories, methodology, andor important applications in quantum chemistry, molecular dynamics, and statistical mechanics. The definitive book on valueatrisk VaR is out in a new second edition, and it is entirely free on this website. Wacom Cte-440 Drivers on this page. Notation and terminology are standardized. Naive Bayes Naive Bayes methods are a set of supervised learning algorithms based on applying Bayes theorem with the naive assumption of independence. Data Mining for Business Analytics Concepts, Techniques, and Applications in XLMiner, Third Edition presents an applied approach to data mining and predictive. Curse Of Dimensionality Pdf' title='Curse Of Dimensionality Pdf' />PDF Work in Progress Liu, Jia, Olivier Toubia, and Shawndra Hill. Content based Temporal Dynamic Model of Web Search Behavior An Application to TV Show Search. Model calibration in progress. Liu, Jia, and Asim Ansari. Understanding Consumer Dynamic Response to Firms Pricing Strategy Under Competing Loyalty Programs. Preparing for submission at Journal of Marketing Research. Liu, Jia, Tejvir Singh Sekhon, Oded Netzer, and Remi Trudel. Social Health Learning Through Crowd Sourced Feedback. Preparing for submission at Journal of Marketing Research.