Automated Fact Repository Construction through Ontology Filtering and Natural Language Processing

In this paper, we present a novel approach to automatic fact repository construction from weblog data. The system we describe uses a combination of ontological Filtering techniques and ontological-semantic natural language processing to collect a concise and thorough view of the data while pro- cessing only a fraction of the blog posts. Ontological Filtering allows our system to weed through each weblog post and select only those relevant to the search topic on hand; the natural language processor extracts meaning representations from the selected texts. These meaning representations are aggregated to produce a collection of machine tractable data concerning the requested search.