tailieunhanh - Báo cáo khoa học: "Chinese Named Entity and Relation Identification System"

In this interactive presentation, a Chinese named entity and relation identification system is demonstrated. The domainspecific system has a three-stage pipeline architecture which includes word segmentation and part-of-speech (POS) tagging, named entity recognition, and named entity relation identitfication. The experimental results have shown that the average F-measure for word segmentation and POS tagging after correcting errors achieves and separately. | Chinese Named Entity and Relation Identification System Tianfang Yao Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai 200030 China yao-tf@ Hans Uszkoreit Department of Computational Linguistics and Phonetics Saarland University Saarbrucken 66041 Germany uszkoreit@ Abstract In this interactive presentation a Chinese named entity and relation identification system is demonstrated. The domainspecific system has a three-stage pipeline architecture which includes word segmentation and part-of-speech POS tagging named entity recognition and named entity relation identitfication. The experimental results have shown that the average F-measure for word segmentation and POS tagging after correcting errors achieves and separately. Moreover the overall average F-measure for 6 kinds of name entities and 14 kinds of named entity relations is and respectively. 1 Introduction The investigation for Chinese information extraction is one of the topics of the project COLLATE DFKI 2002 dedicated to building up the German Competence Center for Language Technology. The presented work aims at investigating automatic identification of Chinese named entities NEs and their relations in a specific domain. Information Extraction IE is an innovative language technology for accurately acquiring crucial information from documents. NE recognition is a fundamental IE task that detects some named constituents in sentences for instance names of persons places organizations dates times and so on. Based on NE recognition the identification of Named Entity Relation NER can indicate the types of semantic relationships between identified NEs. . relationships between person and employed organization person and residing place person and birthday organization and seat etc. The identified results for NEs and NERs can be provided as a resource for other application systems such as questionanswering system. Therefore .