Spatio-temporal models for estimating click-through rate

ABSTRACT

We propose novel spatio-temporal models to estimate click-through rates in the context of content recommendation. We track article CTR at a fixed location over time through a dynamic Gamma-Poisson model and combine information from correlated locations through dynamic linear regressions, significantly improving on per-location model. Our models adjust for user fatigue through an exponential tilt to the first-view CTR (probability of click on first article exposure) that is based only on user-specific repeat-exposure features. We illustrate our approach on data obtained from a module (Today Module) published regularly on Yahoo! Front Page and demonstrate significant improvement over commonly used baseline methods. Large scale simulation experiments to study the performance of our models under different scenarios provide encouraging results. Throughout, all modeling assumptions are validated via rigorous exploratory data analysis.

REFERENCES

• 1: D. Agarwal, B.-C. Chen, P. Elango, and et al. Online models for content optimization. In NIPS, 2008.
• 2: Nicolò Cesa-Bianchi , Paul Fischer, Finite-Time Regret Bounds for the Multiarmed Bandit Problem, Proceedings of the Fifteenth International Conference on Machine Learning, p.100-108, July 24-27, 1998
• 3: Y. Benjamini and Y. Hochberg. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society B, 1995.
• 4: C.A.Colin and P.K.Trivedi. Regression Analysis of Count Data. Cambridge University Press, 1998.
• 5: John M. Chambers, Software for Data Analysis: Programming with R, Springer Publishing Company, Incorporated, 2008
• 6: C.R.Rao. Linear Statistical Inference and Its Applications. Wiley, 2002.
• 7: D.K.Stangl and D.A.Berry. Meta-analysis in Medicine and Health Policy. CRC Press, 2000.
• 8: D.Lambert and C.Liu. Adaptive thresholds: Monitoring streams of network counts. Journal of the American Statistical Association, 2006.
• 9: C. Kaufman, V. Ventura, and R. Kass. Spline-based non-parametric regression for periodic functions and its application to directional tuning of neurons. Statistics in Medicine, 2005.
• 10: Qiaozhu Mei , Kenneth Church, Entropy of search logs: how hard is search? with personalization? with backoff?, Proceedings of the international conference on Web search and web data mining, February 11-12, 2008, Palo Alto, California, USA [doi>10.1145/1341531.1341540]
• 11: Mike West , Jeff Harrison, Bayesian forecasting and dynamic models (2nd ed.), Springer-Verlag New York, Inc., New York, NY, 1997
• 12: Filip Radlinski , Thorsten Joachims, Active exploration for learning rankings from clickthrough data, Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, August 12-15, 2007, San Jose, California, USA [doi>10.1145/1281192.1281254]
• 13: Matthew Richardson , Ewa Dominowska , Robert Ragno, Predicting clicks: estimating the click-through rate for new ads, Proceedings of the 16th international conference on World Wide Web, May 08-12, 2007, Banff, Alberta, Canada [doi>10.1145/1242572.1242643]
• 14: S.P.Ellner and Y.Seifu. Using spatial statistics to select model complexity. Journal of Computational and Graphical Statistics, 2002.
• 15: F. Wu and B. A. Huberman. Novelty and collective attention. In Proceedings of National Academy of Sciences, 2007.
• 16: Fang Wu , Bernardo A. Huberman, Popularity, novelty and attention, Proceedings of the 9th ACM conference on Electronic commerce, July 08-12, 2008, Chicago, Il, USA [doi>10.1145/1386790.1386828]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.5 On-line Information Services

Additional Classification:
  G. Mathematics of Computing
  G.3 PROBABILITY AND STATISTICS

General Terms:
Algorithms, Design, Experimentation, Measurement

Keywords:
content recommendation, ctr positional correlation