A L ENTITY TASK ( MET ) OVERVIEW Roberta Merchant
نویسندگان
چکیده
In November, 1996, the Message Understanding Conference-6 (MUC-6) evaluation of named entity identification demonstrated that systems are approaching human performance on English language texts [10]. Informal and anonymous, the MET provided a new opportunity to assess progress on the same task in Spanish, Japanese, and Chinese. Preliminary results indicate that MET systems in all three languages performed comparably to those of the MUC-6 evaluatien in English.
منابع مشابه
The Multilingual Entity Task (MET) Overview
In November, 1996, the Message Understanding Conference-6 (MUC-6) evaluation of named entity identification demonstrated that systems are approaching human performance on English language texts [10]. Informal and anonymous, the MET provided a new opportunity to assess progress on the same task in Spanish, Japanese, and Chinese. Preliminary results indicate that MET systems in all three language...
متن کاملCerebellar Infarction: Physiotherapeutic Approach – An Overview of Existing Studies
Objectives: Cerebellar infarction is an ischemic type episode involving the three main cerebellar arteries. An infarction in each of these arteries causes different but common symptoms, requiring a particularly important contribution of physiotherapy to its treatment. This overview aimed to investigate the effectiveness of physiotherapy programs on the symptoms of patients with cerebellar infar...
متن کاملNameTag(TM) Japanese and Spanish Systems as Used for MET
We have participated in the Multilingual Entity Task (MET) for Japanese and Spanish using SRA's multilingual text-indexing software called NameTag TM. Its English version was used for the Named Entity Task (NE) in MUC-6 [2]. The NameTag Japanese and Spanish systems were customized to accommodate the MET-specific requirements and were able to achieve high performance in both recall and precision.
متن کاملOverview of TAC-KBP2014 Entity Discovery and Linking Tasks
In this paper we give an overview of the Entity Discovery and Linking tasks at the Knowledge Base Population track at TAC 2014. In this year we introduced a new end-to-end English entity discovery and linking task which requires a system to take raw texts as input, automatically extract entity mentions, link them to a knowledge base, and cluster NIL mentions. In this paper we provide an overvie...
متن کاملOverview of BioNLP Shared Task 2011
The BioNLP Shared Task 2011, an information extraction task held over 6 months up to March 2011, met with community-wide participation, receiving 46 final submissions from 24 teams. Five main tasks and three supporting tasks were arranged, and their results show advances in the state of the art in fine-grained biomedical domain information extraction and demonstrate that extraction methods succ...
متن کامل