BNIE 3.0 (Business news information extraction 3.0)

• Main Authors: Katigbak, Ma. Felina, Oh, Mary Rose, Uy, Joseph Christopher
• Format: text
• Language: English
• Published: Animo Repository 2011
• Subjects: Text processing (Computer science); Natural language processing (Computer science); Computer Sciences
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/11063
• Institution: De La Salle University

Information Extraction (IE) locates and retrieves structured information from unstructured text. Such technology is useful for Business Intelligence (BI), which is the analysis of business data with the help of computers. The research involved the design and implementation of a system that extracts information on business transactions between two or more companies from business news articles and stores these into a database for a more structured view to facilitate comparisons of data. The proponents tested BNIE 3.0 on 75 business news articles form BioSpace.com, and the system was able to achieve an F-measure of 70.28%. With the addition of typed dependencies in semantic interpretation, it has a higher accuracy compared with the 56.27% F-measure of BNIE 2.0 tested on the same 75 business news articles.