Overview of MUC-7/MET-2

Nancy Chinchor

1998 · 159 citations · 0 references

Concepts

TL;DR

The seventh Message Understanding Conference and the Second Multilingual Entity Task involve systems extracting named entities from news text, with examples from the New York Times. The paper describes the tasks performed by systems in MUC‑7 and MET‑2. Named entities were defined as proper names and quantities—including persons, organizations, locations, dates, times, percentages, and monetary amounts—and annotated in SGML within the text stream. The UK satellite broadcaster reported a 17.5 % increase in subscribers to 5.35 million, and the Named Entity task was carried out in Chinese, Japanese, and English.

Abstract

Abstract : The tasks performed by the systems participating in the seventh Message Understanding Conference and the Second Multilingual Entity Task are described here in general terms with examples. On the level of entity extraction, Named Entities (NE) were defined as proper names and quantities of interest. Person, organization, and location names were marked as well as dates, times, percentages, and monetary amounts. The annotation was SGML within the text stream. An example from MUC-7 (New York Times News Service) in English follows. The ENAMEX TYPE=LOCATIONU.K./ENAMEX satellite television broadcaster said its subscriber base grew NUMEX TYPE=PERCENT17.5 percent/NUMEX during TIMEX TYPE=DATEthe past year/TIMEX to 5.35 million The Named Entity task was carried out in Chinese and Japanese (MET-2) concurrently with English (MUC-7).