Kareem S. Aggour, John Interrante, Ibrahim Gokcen
Gathering news events on companies to provide business intelligence to financial investors and creditors is a challenging problem. With a plethora of online news providers and tens of thousands of companies to monitor, automating the extraction and fusion of events is crucial. We developed an intelligent agent-based component framework to query and extract events from multiple providers. This framework integrates multiple machine learning and natural language processing techniques to down-select articles and extract targeted events. Results indicate that our system is capable of extracting focal events on a variety of topics with effective precision and recall.
Subjects: 7.1 Multi-Agent Systems; 1.10 Information Retrieval
Submitted: May 16, 2006