Research

An Overview

What am I doing for my PhD?


I am working with Prof. Anindya Ghose and Prof. Panagiotis G. Ipeirotis in the Center for Digital Economy Research (CeDER). Broadly speaking, my research specialty concentrates on studying the economics of information technology using inter-disciplinary approaches combining econometrics, economic structural modeling, Bayesian modeling and randomized field experiments, with crowd-sourcing and machine learning techniques like text mining, sentiment analysis and image classification. I am most interested in topics related to online social media and search engine marketing. Some examples of reserach questions I am looking at are

- How do consumers search, evaluate and purchase products based on information acquired from search engines and social media?
- How do today's product search mechanisms and social media platforms affect consumers' behavior and the subsequent economic outcome in the market?
- How to leverage knowledge created within and across a large diversity of social media channels, in order to improve the product search mechanism to increase market efficiency and social welfare?
- How can I translate my research into real world applications?

More specifically, my PhD dissertation research concentrates on the following three areas:

First, I empirically demonstrate that using consumer utility theory to build a structural econometric model of demand estimation provides a more precise framework for driving e-commerce search compared to existing models on search engines that typically rely on theories of document relevance and other reduced-form econometric approaches. To achieve the goal, I combine random-coefficient structural model for demand estimation, machine learning techniques such as text mining and image classification, randomized field experiments and crowd-sourcing analysis. On a broader note, the objective of this stage is to illustrate how social media on the Internet can be mined (to learn latent product information) and incorporated into a demand estimation model, and how social media can be leveraged to generate a new ranking system in product search engines to improve the quality of choices available to consumers online. Besides providing consumers with direct economic gains, such a ranking system can lead to non-trivial reduction in consumer search costs and increased usage of product search engines. The first paper from my dissertation is published in Marketing Science. In addition, a related paper won the Best Paper Award at the 20th International World Wide Web Conference (WWW 2011).

Second, I examine how different ranking and personalization mechanisms on product search engines influence consumer online search and purchase behavior. To investigate these effects, I combine archival data analysis with randomized field experiments. In the archival data analysis, I use a hierarchical Bayesian model to jointly estimate the relationship among consumer click and purchase behavior, and search engine ranking decisions. To evaluate the causal effect of search engine interface on user behavior, I conduct randomized field experiments. The field experiments are based on the real-world hotel search engine artifact designed and built by ourselves. By manipulating the default ranking method of search results, and by enabling or disabling a variety of personalization features on the hotel search engine website, I am able to empirically identify the causal impact of search engines on consumers' online click and purchase behavior. A related working paper on this topic will be presented at WISE 2011.

Third, I further examine how consumers search, evaluate and purchase products online. Especially I am interested in consumers' dynamic behavior and strategic actions on product search engines. In particular, this step allows me to more precisely identify consumer preferences not only from the final choices made through the search engines, but also from the entire online search history that leads to the final decisions. Moreover, I am interested to extend this study to a social network setting, and examine how social dynamics may influence consumer search and how social signals can be incorporated into product search engine design. To achieve the goal, I plan to use a dynamic structural model considering customers' forward-looking behavior. The dynamic model enables me to predict the consumer satisfaction in a sense that it extracts the net utility of a product after taking into account the screen position bias, consumer heterogeneity, search cost and social dynamics.

I am going to the ICIS 2011 Doctoral Consortium. For more details on my PhD research, a short version of my dissertation proposal can be found here.


Machine Learning:

Previously, I also worked on projects related to US housing market analysis and forecast, medical image mining for early Alzheimer's Disease diagnose, and blog text mining.


Projects

  WCAI-Wharton Customer Analytics Initiative Research Opportunity for "Examining the Synergetic Effects among Multi-Channel Branding Campaigns." 2012-2013. (with Foster Provost, Ori Stitelman, and Brian Dalessandro)

  MSI-Wharton Interactive Media Initiative Grant for "The Economic Impact of User-Generated Content: Combining Text mining with Demand Estimation in the Hotel Industry." 2009-2010.

  Microsoft Virtual Earth Award., Aug. 2007- May. 2008.

  The EconoMining Project @ NYU, Aug. 2007- 2010.

  Brain image classification for early Alzheimer's Disease detection, Jan 2006-May 2007.

  Blog text mining, Nov 2006-May 2007.


Publications

Designing Ranking Systems for Hotels on Travel Search Engines by Mining User-Generated and Crowd-Sourced Content. Marketing Science. 2012. (with Anindya Ghose and Panagiotis G. Ipeirotis)

Towards a Theory Model for Product Search. Proceedings of the 20th International World Wide Web Conference (WWW 2011), Hyderabad, India, March 2011. Best Paper Award, (Slides, A Short Version for Demo). (with Anindya Ghose and Panagiotis G. Ipeirotis)

Designing Ranking Systems for Hotels on Travel Search Engines to Enhance User Experience. Proceedings of the International Conference on Information Systems (ICIS 2010), St. Louis, December 2010. (with Anindya Ghose and Panagiotis G. Ipeirotis)

Extracting Economic Value from Online User-Generated Content: Combining Text Mining with Demand Estimation in the Hotel Industry. In Proceedings of the 32nd INFORMS Marketing Science Conference, Cologne, Germany, June 2010. (with Anindya Ghose and Panagiotis G. Ipeirotis)

Enhancing Clustering Blog Documents by Utilizing Author/Reader Comments. In Proceedings of the 45th ACM Southeast Conference (ACMSE 2007), pp.94-99, March 23-24, 2007, Winston-Sale, North Carolina (with Shuting Xu and Jun Zhang)


Workshops

 

 Examining the Impact of Search Engine Ranking and Personalization on Consumer Behavior: Combining Bayesian Modeling with Randomized Field Experiments. Workshop on Information Systems and Economics (WISE 2011), Shanghai, December, 2011. (With Anindya Ghose and Panagiotis G. Ipeirotis)

 Mining User-Generated and Crowd-Sourced Content on the Internet to Estimate Demand for Hotels. The Sixth Symposium on Statistical Challenges in Electronic Commerce Research (SCECR 2010). Austin, TX, June 2010. (With Anindya Ghose and Panagiotis G. Ipeirotis)

 Improving Product Search with Economic Theory. ICDE 2010 PhD Workshop In conjunction with ICDE 2010, Long Beach, CA, March 2010. (Receive $1000 fellowship by NSF grant IIS-0956600.)

 The Economic Impact of User-Generated Content on the Internet: Combining Text Mining with Demand Estimation in the Hotel Industry.. Workshop on Information Systems and Economics (WISE 2009), Phoenix, December, 2009. (with Anindya Ghose and Panagiotis G. Ipeirotis)

 Towards Designing Ranking Systems for Hotels on Travel Search Engines: Combining Text mining with Demand Estimation in the Hotel Industry. Workshop on Information Technology and Systems (WITS 2009), Phoenix, December, 2009. (with Anindya Ghose and Panagiotis G. Ipeirotis)

Stay Elsewhere? Improving Local Search for Hotels Using Econometric Modeling and Image Classification (WebDB 2008, in conjunction with ACM SIGMOD/PODS 2008, Vancouver, Canada. (with Anindya Ghose and Panagiotis G. Ipeirotis)

Improving Local Search for Hotels Using Econometric Modeling and Image Classification. Microsoft Virtual Earth and Location Summit, Redmond, WA, May, 2008. (with Anindya Ghose and Panagiotis G. Ipeirotis)

Stay Elsewhere? The Economic Impact of Location-based Hotel Features: A View from Remote Sensing Image Analysis. Winter Conference on Business Intelligence, March, 2008, Salt Lake City, Utah. (with Anindya Ghose and Panagiotis G. Ipeirotis)


Awards

 WCAI-Wharton Customer Analytics Initiative Research Opportunity for "Examining the Synergetic Effects among Multi-Channel Branding Campaigns." 2012-2013. (with Foster Provost, Ori Stitelman, and Brian Dalessandro)

 ICIS Doctoral Consortium 2011, invited participant, Shanghai, China. Dec. 2011.

 Best Paper Award, World Wide Web Conference 2011(WWW2011), Hyderabad, India.

Invited participant for the Columbia-Duke-UCLA Workshop on Quantitative Marketing and Structural Econometrics (QuantCamp). Duke University, Auguest, 2010.

 PhD Student Travel Support Fellowship, Stern School of Business, NYU, 2009-2010.

 Doctoral Student Fellowship, Leonard N. Stern School of Business, New York University, Sept. 2008-present.

 Student Travel Support from Graduate School Fellowship of University of Kentucky, Spring, 2007.

 Best Paper Award, 20th Annual EKU Symposium in the Mathematical, Statistical and Computer Sciences, Mar 31, 2006.

 Editor's Choice Award for Outstanding Achievement in Poetry, International Library of Poetry, July 2006.

 Seventeen other Fellowships and Awards in my B.S/M.S study in China, 1999-2005.


Talks, Demonstrations, and others

 Towards a Theory Model for Product Search. Baidu Inc. Beijing, China, June 2011.

 A Demo Search Engine for Products. WWW 2011. March 2011.

 Designing Ranking Systems for Hotels on Travel Search Engines by Mining User-Generated and Crowd-Sourced Content. 6th bi-annual Conference on The Economics of Intellectual Property, Software and the Internet. Toulouse, France, January 2011.

 A Demo for Product Search. Day 1: Technology Expo, NYU. November 2010.

 Designing Ranking Systems for Hotels on Travel Search Engines by Mining User-Generated and Crowd-Sourced Content. The 10th Anniversary for Department of Information, Operations & Management Science (IOMS), Stern School of Business, NYU. October 29th, 2010.

 Designing Ranking Systems for Hotels on Travel Search Engines. New York Computer Science and Economics Day. The New York Academy of Sciences, World Trade Center, New York, NY. 2009, 2010.

 Stay Elsewhere? Improving Local Search for Hotels Using Econometric Modeling and Image Classification WebDB 2008, in conjunction with ACM SIGMOD/PODS 2008 , Vancouver, Canada.

 Local Search for Hotel and Restaurants Using Econometrics Modeling and Image Classification. Virtual Earth Academic Research Collaboration 2007 RFP Awards Summits. April 30 - May 1, 2008. Redmond, Washington.

Stay Elsewhere? The Economic Impact of Location-based Hotel Features: A View from Remote Sensing Image Analysis. Winter Conference on Business Intelligence, March 20-22, 2008, Salt Lake City, Utah.

Enhancing Clustering Blog Documents by Utilizing Author/Reader Comments. Mar. 2007. North Carolina.


Bookmarks

< Econometric & Statistical Data Analysis>

TOOLs

STATA

Matlab

Resources

UCLA Stat Computing

Online Discussion Panel

Econ-PhDs.net

PhD in Economics Forum

< Others >

Unix & Solaris

Solaris 8----User commands

Unix programming

Unix Power

UNIX-C System Interfaces & Headers

LaTex

LaTex Help

Hypertext Help with LaTeX

LaTeX Symbols