A Markov chain model for integrating behavioral targeting into contextual advertising

ABSTRACT

Both Contextual Advertising (CA) and Behavioral Targeting (BT) are playing important roles in online advertising market. Recently, the problem of how to integrate BT strategies into CA has attracted much attention from both industry and academia. However, to our best knowledge, few research works have been published to provide BT solutions in CA. In this paper, we propose a new notion of relevance between webpages and ads based on users' online click-through behaviors from BT's perspective. Compared with the classical behavior targeting method where only users' history interests are considered, we pay more attention to the click probability of ads from a webpage where the relevance between them is evaluated. Moreover, a combination model integrating behavioral relevance and contextual relevance for matching ads and webpags is presented. The model parameters are learnt from a dataset consisting of 200 webpages and 35,880 ads. Experimental results show that our integrated strategy indeed outperforms the strategies that only consider either behavioral relevance or contextual relevance. The best model achieves a 18.1% improvement in precision over single strategies.

REFERENCES

• 1: Ricardo A. Baeza-Yates , Berthier Ribeiro-Neto, Modern Information Retrieval, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1999
• 2: J. S. Breese, D. Heckerman, and C. Kadie. Empirical analysis of predictive algorithms for collaborative filtering. In UAI, pages 43--52, 1998.
• 3: Andrei Broder , Marcus Fontoura , Vanja Josifovski , Lance Riedel, A semantic approach to contextual advertising, Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval, July 23-27, 2007, Amsterdam, The Netherlands [doi>10.1145/1277741.1277837]
• 4: Patrali Chatterjee , Donna L. Hoffman , Thomas P. Novak, Modeling the Clickstream: Implications for Web-Based Advertising Efforts, Marketing Science, v.22 n.4, p.520-541, October 2003 [doi>10.1287/mksc.22.4.520.24906]
• 5: D. C. Fain and J. Pedersen. Sponsored Search: A Brief History. Bulletin of the American Society for Information Science and Technology, 2005.
• 6: D. Hawking, N. Craswell, and P. B. Thistlewaite. Overview of TREC-7 very large collection track. In The Seventh Text Retrieval Conference (TREC-7), pages 91--104, Gaithersburg, Maryland, USA, November 1998.
• 7: R. E. Marshall, D. A. de Wolf, C. Kontogeorgakis, Absorption and Reflection in Atmospheric Haze at Visible Through Far Infrared Wavelengths. Technical Memorandum, RTI/4500/046--04S, 1996.
• 8: R. H. J. M. Otten , L. P. P. P. van Ginneken, The annealing algorithm, Kluwer, B.V., Deventer, The Netherlands, 1989
• 9: Paul Resnick , Neophytos Iacovou , Mitesh Suchak , Peter Bergstrom , John Riedl, GroupLens: an open architecture for collaborative filtering of netnews, Proceedings of the 1994 ACM conference on Computer supported cooperative work, p.175-186, October 22-26, 1994, Chapel Hill, North Carolina, United States [doi>10.1145/192844.192905]
• 10: Berthier Ribeiro-Neto , Marco Cristo , Paulo B. Golgher , Edleno Silva de Moura, Impedance coupling in content-targeted advertising, Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval, August 15-19, 2005, Salvador, Brazil [doi>10.1145/1076034.1076119]
• 11: Sheldon M. Ross, Introduction to Probability Models, Ninth Edition, Academic Press, Inc., Orlando, FL, 2006
• 12: Ryen W. White , Mikhail Bilenko , Silviu Cucerzan, Studying the use of popular destinations to enhance web search interaction, Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval, July 23-27, 2007, Amsterdam, The Netherlands [doi>10.1145/1277741.1277771]
• 13: Jun Yan , Ning Liu , Gang Wang , Wen Zhang , Yun Jiang , Zheng Chen, How much can behavioral targeting help online advertising?, Proceedings of the 18th international conference on World wide web, April 20-24, 2009, Madrid, Spain [doi>10.1145/1526709.1526745]
• 14: Wen-tau Yih , Joshua Goodman , Vitor R. Carvalho, Finding advertising keywords on web pages, Proceedings of the 15th international conference on World Wide Web, May 23-26, 2006, Edinburgh, Scotland [doi>10.1145/1135777.1135813]
• 15: Adlink. https://www.google.com/adsense/login/en_US/?gsessionid =Dc28hZShnCI
• 16: AlmondNet. http://www.almondnet.com/
• 17: BlueLithium. http://www.bluelithium.com/
• 18: Burst. http://www.burstmedia.com/
• 19: DoubleClick. http://www.doubleclick.com/products/dfa/index.aspx
• 20: eMarketer. http://www.emarketer.com/Article.aspx?id=1006813
• 21: Google AdSense. http://www.newsoxy.com/technology/google-adsense/article11725.html
• 22: NebuAd. http://www.nebuad.com/
• 23: Phorm. http://www.phorm.com/
• 24: Revenue Science. http://www.revenuescience.com/advertisers/advertiser_solutions.asp
• 25: Specificmeida. http://www.specificmedia.co.uk/
• 26: TACODA. http://www.tacoda.com/
• 27: Yahoo! Smart Ads. http://advertising.yahoo.com/marketing/smartads/

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications
          Subjects: Data mining

Additional Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval

General Terms:
Algorithms, Experimentation

Keywords:
Markov chain model, behavioral targeting, click-through data, contextual advertising