Using structured self-organizing maps in news integration websites

• Main Authors: Perelomov, Ivan, Azcarraga, Arnulfo P., Tan, Jonathan, Chua, Tat Seng
• Format: text
• Published: Animo Repository 2007
• Subjects: Self-organizing maps; News Web sites; Data integration (Computer science); Computer Sciences
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/11961
• Institution: De La Salle University

The Bveritas system integrates and organizes news articles from English news websites based in Singapore, Malaysia, Philippines, and Thailand, plus news stories from CNN and Reuters International. The main features of the system are the following: 1) automatic clustering of news documents into themes; 2) ordering of these news clusters in a theme map; 3) extraction of meaningful labels for each cluster of news articles; 4) use of extracted labels and title words to retrieve ranked news article lists based on query words; 5) summarization of news articles; and 6) automatic generation of links to related news articles. The novel features of the system result directly from the use of a structured self-organizing map as the underlying logical structure for the news archive.