A new sequential classification to assist Ad auction agent in making decisions

One of the critical abilities of intelligent agents is making quick, accurate and wise decision within dynamic environments in a reasonable period of time. This paper introduces a new sequence classification method based on positive and negative sequential patterns. The historical log data from previous performance “TAC/Ad auction” has been used for the classification. We have applied Nconf: an “interestingness measure” to prune meaningless negative and positive candidates. Our experiments showed that agents equipped with this classifier can earn higher profit compared to non equipped ones. Our findings lead us to believe that with some additional tune during the run we may achieve more qualified agents.

[1]  Frank Dignum,et al.  Towards socially sophisticated BDI agents , 2000, Proceedings Fourth International Conference on MultiAgent Systems.

[2]  Pericles A. Mitkas,et al.  Data mining for agent reasoning: A synergy for training intelligent agents , 2007, Eng. Appl. Artif. Intell..

[3]  R. Sreerama Kumar,et al.  Intelligent decision making in multi-agent robot soccer system through compounded artificial neural networks , 2007, Robotics Auton. Syst..

[4]  Qiming Chen,et al.  PrefixSpan,: mining sequential patterns efficiently by prefix-projected pattern growth , 2001, Proceedings 17th International Conference on Data Engineering.

[5]  He Jiang,et al.  Research on Mining Sequential Positive and Negative Association Rules , 2009, 2009 Second International Conference on Intelligent Computation Technology and Automation.

[6]  Chengqi Zhang,et al.  Efficient Mining of Event-Oriented Negative Sequential Rules , 2008, 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology.

[7]  Wei-Hua Hao,et al.  Mining negative sequential patterns , 2007 .

[8]  Pericles A. Mitkas,et al.  A retraining methodology for enhancing agent intelligence , 2005 .

[9]  Maria E. Orlowska,et al.  Finding Negative Event-Oriented Patterns in Long Temporal Sequences , 2004, PAKDD.

[10]  Andreas D. Lattner,et al.  Sequential Pattern Mining for Situation and Behavior Prediction in Simulated Robotic Soccer , 2005, RoboCup.

[11]  Zili Zhang,et al.  Agents and Data Mining: Mutual Enhancement by Integration , 2005, AIS-ADM.

[12]  Xiaoping Fan,et al.  A New Interestingness Measure of Association Rules , 2008, 2008 Second International Conference on Genetic and Evolutionary Computing.

[13]  Henryk Rybinski,et al.  Knowledge sharing in default reasoning based multi-agent systems , 2003, IEEE/WIC International Conference on Intelligent Agent Technology, 2003. IAT 2003..

[14]  Pericles A. Mitkas,et al.  A Methodology for Predicting Agent Behavior by the Use of Data Mining Techniques , 2005, AIS-ADM.

[15]  Chengqi Zhang,et al.  Agent-Mining Interaction: An Emerging Area , 2007, AIS-ADM.

[16]  Gregory Piatetsky-Shapiro,et al.  Discovery, Analysis, and Presentation of Strong Rules , 1991, Knowledge Discovery in Databases.

[17]  Engelbert Mephu Nguifo,et al.  Improving the Behavior of Intelligent Tutoring Agents with Data Mining , 2009, IEEE Intelligent Systems.

[18]  Ming-Yen Lin,et al.  Mining Negative Sequential Patterns for E-commerce Recommendations , 2008, 2008 IEEE Asia-Pacific Services Computing Conference.

[19]  Wil M. P. van der Aalst,et al.  Analyzing Multi-agent Activity Logs Using Process Mining Techniques , 2008, DARS.

[20]  Wei-Min Ouyang,et al.  Mining Negative Sequential Patterns in Transaction Databases , 2007, 2007 International Conference on Machine Learning and Cybernetics.

[21]  Ramakrishnan Srikant,et al.  Mining Sequential Patterns: Generalizations and Performance Improvements , 1996, EDBT.