Analysis of the query logs of a Web site search engine

© 2005 Wiley Periodicals, Inc. • Published online 31 August 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/asi.20210 that automatically collect Web pages and create an index that can be searched by users (Chau & Chen, 2003b). As these general-purpose search engines do not restrict themselves to particular domains or specialties, they often try to collect as many Web pages as possible. However, as the number of indexable pages on the Web has exceeded 3 billion, it has become more difficult for these search engines to keep an up-to-date and comprehensive search index; low precision and low recall rates often result. On the other hand, many Web sites have their own search engines. A Web site search engine is one that allows users to search for pages only within a particular Web domain or Web host. It is often found in the main page of a Web site and only indexes pages in that particular Web site. For example, going to the homepage of Microsoft Corporation (http://www.microsoft.com/), one would see a text box on the page that allows users to perform a search restricted to the microsoft.com Web site. Because a Web site search engine only needs to work with a limited set of pages, it can provide customized features to users and can be updated much more frequently (e.g., daily or even hourly) than general-purpose search engines. In addition, a Web site search engine is often more comprehensive because it can index pages that are not accessible by search engine crawlers/spiders, and no general-purpose search engines can cover every single page in every single Web site on the Internet. Although Web site search engines are very useful for users looking for information on a particular Web site, these search engines often have different user interfaces, interpret queries in different ways, support different types of advanced search functionalities, and employ different search algorithms. From end users’ point of view, dealing with an array of different interfaces and understanding each one’s idiosyncrasies add much confusion and present an additional layer of information and cognitive overload. Most users find it difficult to adapt to different search engines and cannot Analysis of the Query Logs of a Web Site Search Engine