Novel Text Summarization Techniques for Contextual Advertising

Novel Text Summarization Techniques for Contextual Advertising

Giuliano Armano (University of Cagliari, Italy) and Alessandro Giuliani (University of Cagliari, Italy)
DOI: 10.4018/978-1-4666-5019-0.ch008
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Recently, there has been a renewed interest on automatic text summarization techniques. The Internet has caused a continuous growth of information overload, focusing the attention on retrieval and filtering needs. Since digitally stored information is more and more available, users need suitable tools able to select, filter, and extract only relevant information. This chapter concentrates on studying and developing techniques for summarizing Webpages. In particular, the focus is the field of contextual advertising, the task of automatically suggesting ads within the content of a generic Webpage. Several novel text summarization techniques are proposed, comparing them with state of the art techniques and assessing whether the proposed techniques can be successfully applied to contextual advertising. Comparative experimental results are also reported and discussed. Results highlight the improvements of the proposals with respect to well-known text summarization techniques.
Chapter Preview
Top

Background

This section is aimed at giving a summary about TS and CA, recalling the main contributions in these fields. Then, the section provides the reasons why TS should be performed by a CA system. Finally, a generic architecture for CA is illustrated, describing also the baseline system to be used as starting point for experiments.

Complete Chapter List

Search this Book:
Reset