LINGUIST List 15.2698

Wed Sep 29 2004

Media: Web Content Analysis -- US Election 2004

Editor for this issue: Jessica Boynton <>


        1.    Arno Scharl, Web Content Analysis -- US Election 2004

Message 1: Web Content Analysis -- US Election 2004

Date: 28-Sep-2004
From: Arno Scharl <>
Subject: Web Content Analysis -- US Election 2004

Dear Colleagues,

Five weeks prior to the US Election 2004, the ECOresearch Network has launched a Web monitoring system to track the candidates' performance on the campaign trail ( Weekly snapshots of media coverage showcase the potential of automated content analysis by revealing regional differences and contrasting the sites of international media, the Fortune 1000 and environmental organizations.

Analyzing these sites yields more than 500,000 documents each week, comprising about 125 million words in 11 million sentences. An automated process then identifies attention by counting the number of references to a candidate. It measures attitude by associating these references with positive and negative terms. Keywords, grouped by political party and geographic region, reflect topics associated with the candidates.

Best regards, Arno Scharl

DDr Arno Scharl
Professor of Information Systems
University of Western Australia, Business School
35 Stirling Highway, Crawley, WA 6009, Australia

Linguistic Field(s): Applied Linguistics; Computational Linguistics; Text/Corpus Linguistics

Respond to list|Read more issues|LINGUIST home page|Top of issue