Searching multimedia federated content web collections

Abstract

Purpose

The purpose of this research is to show that federated content collections are important for providing access to multiple content repositories, including image, video, audio and Web sites.

Design/methodology/approach

This paper presents findings from an analysis of differences in users' Web searching patterns as they access various federated content collections. A dataset of 4,056,374 records submitted to the Dogpile.com Web meta‐search engine were analysed. An analysis was conducted of search session length, query length, number of results pages viewed, use of systems' assistance and the frequency of repeat queries.

Findings

Overall, users entered two to three terms per query and examined only the first pages of results. However, findings include differences in users' access patterns to various content collections. Web, news and audio queries were longer sessions but shorter queries. More users seeking images and videos sought systems assistance.

Originality/value

This is a large‐scale original study using data from a commercial Web search engine. The paper provides a valuable comparison of different types of search – text v. audio, image, etc.

Citation

Spink, A. and Jansen, B.J. (2006), "Searching multimedia federated content web collections", Online Information Review, Vol. 30 No. 5, pp. 485-495. https://doi.org/10.1108/14684520610706389