Optimization of restricted searches in web directories using hybrid data structures

5Citations
Citations of this article
23Readers
Mendeley users who have this article in their library.
Get full text

Abstract

The need of efficient tools in order to manage, retrieve and filter the information in the WWW is clear. Web directories are taxonomies for the classification of Web documents. These kind of information retrieval systems present a specific type of search where the document collection is restricted to one area of the category graph. This paper introduces a specific data architecture for Web directories that improves the performance of restricted searches. That architecture is based on a hybrid data structure composed of an inverted file with multiple embedded signature files. Two variants are presented: hybrid architecture with total information and with partial information. This architecture has been analyzed by means of developing both variants to be compared with a basic model. The performance of the restricted queries was clearly improved, especially the hybrid model with partial information, which yielded a positive response under any load of the search system. © Springer-Verlag Berlin Heidelberg 2003.

Cite

CITATION STYLE

APA

Cacheda, F., Carneiro, V., Guerrero, C., & Viña, A. (2003). Optimization of restricted searches in web directories using hybrid data structures. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2633, 436–451. https://doi.org/10.1007/3-540-36618-0_31

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free