A Ranking Method for Relaxed Queries in Book Search
by
Momo Kyozuka,
Yang Xu,
Keishi Tajima
Abstract
In this paper, we propose a ranking method for keyword-based book
search systems. A user issues a query consisting of keywords
describing the contents of the book, and the system returns a ranked
list of candidate books. Because we do not have full text data of all
the books, we use a database of brief descriptions of books in the
market currently or in the past. When such brief descriptions are
only available, some query keywords may not appear in the description
of the book the user is looking for. To solve that problem, our
method ranks books in two steps. We first generate relaxed queries by
removing some keywords from the given original query,
and rank them based on how likely the remaining keywords appear in the
brief descriptions. We then retrieve matching books for each query,
find words in the description that are the most similar to the removed
keywords, and rank the books based on that similarity. By combining
these two rankings, i.e., the ranking of relaxed queries, and the
ranking of books matching with each query, we produce the final
ranking. In this paper, we focus on the ranking method for the second
step. Our experiment shows that our method is effective when the
original query includes many keywords that do not appear in the
description of the target book.
/ Fax: +81(Japan) 75-753-5978 / Office: Research Bldg. #7, room 404