Automatic Query-Focused Summary Generation System for Tourism Discourse Using Rhetorical Structure Theory: Cognitive and Multimodal Approach to Tourism Discourse

Abstract

Summary generation systems when integrated with Information Retrieval (IR) can give an idea about the retrieved web pages to the user before even the user opens the web page. The summary could be generated for a single web page or for a set of web pages retrieved for a given query. When such a system is built for tourism web sites, the user can get a summary of a particular tourist spot or about the tourist spots present in a particular place. This chapter describes about such a summary generation system which is built using Rhetorical Structure Theory (RST). RST is a well-known discourse theory which is used for discourse analysis of text documents. The RST makes use of another semantic representation namely, Universal Networking Language (UNL) to find the coherent text fragments. These coherent text fragments are indexed and linked with an IR system. When a user gives a query, the web pages along with a single document and multi document summary.