论文信息 - Planning of interactive information retrieval by means of reinforcement learning

Planning of interactive information retrieval by means of reinforcement learning

The theoretical base of a new approach to automation of interactive information retrieval on the basis of reinforcement learning methods is considered. Advantages and problem questions of the use of reinforcement learning for this task are analyzed and the peculiarities of the task are defined

A. Sachenko | V. Vitsentiy | A. Spink

[1] Richard S. Sutton,et al. Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[2] Dayne Freitag,et al. A Machine Learning Architecture for Optimizing Web Search Engines , 1999 .

[3] V. Vitsentiy. Improvement of human-machine interaction with applications to information retrieval system , 2002, Proceedings First International IEEE Symposium Intelligent Systems.

[4] Andrew W. Moore,et al. Generalization in Reinforcement Learning: Safely Approximating the Value Function , 1994, NIPS.

[5] Andrew McCallum,et al. Using Reinforcement Learning to Spider the Web Efficiently , 1999, ICML.

[6] Andrew W. Moore,et al. Reinforcement Learning: A Survey , 1996, J. Artif. Intell. Res..