On Refining Real-time Multilingual News Event Extraction through Deployment of Cross-Lingual Information Fusion Techniques

Abstract

Nowadays, many influential security-related facts are reported multiple times by different sources and in dif- ferent languages. Therefore, in the recent years, the research on advancing event extraction technology shifted from classical single-document extraction toward cross-document information aggragation and fact validation. However, relatively little work has been reported on cross-lingual information fusion in this area. This paper presents the results of some preliminary experi- ments on deploying cross-lingual information fusion techniques for refining the results of a large-scale multilingual news event extraction system. Circa 8% of event descriptions extracted by the mono-lingual systems could be refined. The overall gain in recall and precision against the best mono-lingual system was 6,4% and 4,8% resp.