Evolutionary Prediction of Online Keywords Bidding

Authors:
Liwen Hou;Liping Wang;Jinggang Yang
Affiliations:
Department of Management Information System, Shanghai Jiaotong University, Shanghai, China;Department of Management Information System, Shanghai Jiaotong University, Shanghai, China;Department of Management Information System, Shanghai Jiaotong University, Shanghai, China
Venue:
EC-Web '08 Proceedings of the 9th international conference on E-Commerce and Web Technologies
Year:
2008

Citing 2
Cited 1

Bayesian Networks and Decision Graphs

Bayesian Networks and Decision Graphs
Use of Markov chains to design an agent bidding strategy for continuous double auctions

Journal of Artificial Intelligence Research

Maturity, distance and density (MD2) metrics for optimizing trust prediction for business intelligence

Journal of Global Optimization

Quantified Score

Hi-index	0.00

Visualization

Abstract

Online keywords bidding as a new business model for search engine market facilitates the prosperity of internet economy as well as attracts myriad small business to directly target their customers. In order to effectively manage each advertising campaign the manager needs to figure out smart strategy to bid. This paper successively developed two models (static and dynamic) with four crucial variants at keyword bidding (bid, rank, click through and impression) based on the bayesian network to assist the prediction of bidding. Herein dynamic model, evolved from the static model, takes the influence of click through on the last period into account and extends the strategy space. Empirical study by aid of data from the largest online travel agency in China is carried out to test both models and the results indicate that they are effective while the dynamic model is more attractive in terms of prediction accuracy. Finally further research directions along this paper is shown.