Method to Reduce Complexity and Response Time in a Web Search

Method to Reduce Complexity and Response Time in a Web Search

María R. Romagnano, Silvana V. Aciar, Martín G. Marchetta
Copyright: © 2018 |Pages: 17
DOI: 10.4018/978-1-5225-5191-1.ch006
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Living in times of technological changes that alter our daily activities, involving tasks such as reading the newspaper, following the weather, scheduling a trip, are usually executed after perusal of the gigantic repository of information, commonly known as the World Wide Web. However some problems are still associated with the information found in such a vast amount of information: heterogeneity, availability, distribution, quality and quantity of irrelevant information. Recent work has suggested different ways of grouping similar information sources, trying to give solutions to these problems. However, some domains are more complex than others. For example, a person looking for tourist information, is generally overwhelmed by visiting various websites. This paper proposes the implementation of a method to retrieve and group web information sources, depending on the services they offer; thereby allowing the user to get accurate answers; thus reducing the time and complexity in the search.
Chapter Preview
Top

Bibliographic Review

The IR (Information Retrieval) helps users to find relevant and necessary information from a large collection of text documents. Technically, the IR studies the acquisition, organization, storage, retrieval and distribution of information (Liu 2007, pp.183-235).

The work of Han and Kamber (2006, p. 615) mentions that information retrieval is the area dedicated to the organization and retrieval of information from a large amount of text-based documents. Although these have been developed in parallel with database systems; they differ from the latter in some presented problems, such as de-structuring of documents, searchable by approximation based on keywords and the notion of relevance.

To carry out a subsequent analysis of the retrieved information, web pages can be grouped according to a specific criterion. This grouping can be performed using supervised learning techniques such as classification; or unsupervised learning techniques such as clustering (Baldi, Frasconi, & Smyth 2003, p. 17).

Complete Chapter List

Search this Book:
Reset