A Methodology for IR Query Clarification based on Adaptive Generalization

摘要

We can use various types of information retrieval (IR) system to access large quantities of text data. These systems usually request users to input keywords to represent their retrieval intention. Since, it is difficult for a user to select appropriate keywords, many IR systems have the capability of modifying keywords by estimating the user's intention. However, because modified keywords are usually represented in a complicated form, it is difficult to judge how appropriate they are. In this research, we assume all relevant documents should have words that correspond to concepts that the user would like to retrieve. Therefore, a boolean query formula is one good way to represent user's intent with higher readability. In order to find out these concepts in a relevant document set, we have already proposed a new IR system that calculates appropriate generalized concept to represent user's intention. In this study, we implement a user interface that supports to clarify IR query by using abstract concepts.