Paper: Bootstrapping For Named Entity Tagging Using Concept-Based Seeds

ACL ID N03-2025
Title Bootstrapping For Named Entity Tagging Using Concept-Based Seeds
Venue Human Language Technologies
Session Short Paper
Year 2003
Authors

A novel bootstrapping approach to Named Entity (NE)tagging using con- cept-based seeds and successive learners is presented. This approach only requires a few common noun or pronoun seeds that correspond to the concept for the tar- geted NE, e.g. he/she/man/woman for PERSON NE. The bootstrapping proce- dure is implemented as training two suc- cessive learners. First, decision list is used to learn the parsing-based NE rules. Then, a Hidden Markov Model is trained on a corpus automatically tagged by the first learner. The resulting NE system ap- proaches supervised NE performance for some NE types.