Master Thesis - 2010

Recently, as networks widely spread, the number of contents over networks is increasing. For efficient and flexible information retrieval over such networks, agent technology receives much attention. There is the case where a retrieved information has difference of quality by user's request and can be given scores. We proposed an agent execution control method for time-constrained information retrieval. It achieves to find better results by termination of an agent which already acquires results with enough quality or has less probability to improve the quality by continuing retrieving.

In this method, however, it is supposed that the kind of score distribution is known accurately. Though, to know the score distribution is very difficult and not realistic. In this paper, I suppose that the score distribution is known but its parameters are not known. Then I propose the execution control method using the parameters which are estimated during retrieval. In order to make it effective, I evaluate the number of data to start estimation and the effect of inaccurate estimation.

Finally, the performance of the proposed method is evaluated by simulation experi- ments. I confirmed that the execution control method with estimation is superior to compared method.