CMIC at INEX 2007: Book Search Track

摘要

With massive book digitization efforts underway, the need for effective retrieval of books and pages in books is an important problem. This paper describes our submissions to the INEX 2007 Book Search track. We explored using book specific features such as table of content and index pages and headers along with non-book specific features. Our results show that indexing the entire contents of books and headers provided the most effective retrieval strategy.