Exploring the Usefulness of Cross-lingual Information Fusion for Refining Real-time News Event Extraction: A Preliminary Study
Jakub Piskorski, Jenya Belayeva, Martin Atkinson · 2011
Nowadays, many influential facts are reported multiple times by different sources and in different languages. This paper presents the results of an experiment on deploying cross-lingual information fusion techniques for refining the results of a large-scale multilingual news event extraction system. An evaluation on a test corpus consisting of 618 event descriptions which refer to 523 real-world events revealed that the description of circa 10 % of the events extracted by the mono-lingual systems could be refined. In particular, an overall gain of 6,4 % and 4,8 % in recall and precision against the best monolingual system could be obtained respectively. 1