Abstract: Resource selection is one of the most important problems in distributed information retrieval. In this paper, resource selection is converted to document retrieval problem. Some document retrieval method is adopted to solve the resource selection problem. Each retrieval method is tested both on the document retrieval and resource selection problem. Compared to the well known resource selection method CORI, language model based retrieval method achieves a quite promising result.

Key words: Distributed information retrieval, Resource selection, Language model

摘要： 集合选择是分布式信息检索中的重要问题，将集合选择问题转化为文档检索问题，尝试了多种文档检索方法来解决集合选择问题，并将各种方法的文档检索结果与集合选择结果进行了对比，通过与经典的集合选择算法CORI相比较，实验发现语言模型的集合选择方法能够取得令人满意的结果。

关键词: 分布式信息检索, 集合选择, 语言模型