A model of cluster searching based on classification

摘要

The use of document clusters has been suggested as an efficient file organization for a document retrieval system. It is possible that by using this information about the relationships between documents that the effectiveness of the system (i.e. its ability to distinguish relevant from non-relevant documents) may also be improved. In this paper a probabilistic model of cluster searching based on query classification is described. This model is tested with retrieval experiments which indicate that it can be more effective than heuristic cluster searches and cluster searches based on other models. It can also be more effective than a full search in which every document is compared to the query. The efficiency aspects of the implementation of the model are discussed.