Computing Reviews

Liquid query:multi-domain exploratory search on the Web
Bozzon A., Brambilla M., Ceri S., Fraternali P.  WWW 2010 (Proceedings of the 19th International Conference on the World Wide Web, Raleigh, NC, Apr 26-30, 2010)161-170,2010.Type:Proceedings
Date Reviewed: 08/25/10

The huge amount of information on the Web and the nature of human languages (and thus, the queries) make general-purpose search engines less desirable. At the same time, domain-specific search engines generally restrict their search results within the designated domains. Bozzon et al.’s liquid query project tackles this general search problem using multi-domain exploratory search. This paper describes the concept, the general architecture, and an example of multi-domain search that is supported by the liquid query approach.

The general idea of the multi-domain exploratory search is to process the original user query, send it to individual search engines in different domains, collate the search results, and present them to the users. Specifically, the liquid query life cycle consists of four phases: application configuration, where the designers define a liquid query template for a specific application, such as travel; query submission, where the user submits the query in compliance with the template; query execution, where the liquid query system sends the query to multiple search engines and collates the search results based on the user preference; and result browsing, where the user is able to manipulate the search results using the interface provided by the liquid query system. The example system described in the paper uses the Yahoo! query language (YQL) database as the backend. Thus, the search results are obtained and manipulated using YQL--similar to any standard structured query language (SQL).

The authors’ ideas and experimental system present an interesting and effective way of combating search inaccuracy due to information overflow. Their multi-domain search approach seems to parallel what metasearch engines did in the early days of search engines; it has great potential.

Reviewer:  Xiannong Meng Review #: CR138328 (1107-0766)

Reproduction in whole or in part without permission is prohibited.   Copyright 2024 ComputingReviews.com™
Terms of Use
| Privacy Policy