A bootstrapping method for learning semantic lexicons using extraction pattern contexts
EMNLP '02 Proceedings of the ACL-02 conference on Empirical methods in natural language processing - Volume 10
Unsupervised named-entity extraction from the Web: An experimental study
Artificial Intelligence
Unsupervised named-entity recognition: generating gazetteers and resolving ambiguity
AI'06 Proceedings of the 19th international conference on Advances in Artificial Intelligence: Canadian Society for Computational Studies of Intelligence
Hi-index | 0.00 |
Named entities are important content-carrying units within documents. Consequently named entity recognition (NER) is an important part of information extraction. One fast and accurate approach to NER uses a list or gazette consisting of known instances. Gazette creation problem considers how to automatically create a comprehensive gazette from given unlabeled document repository. We describe an unsupervised algorithm for automatic gazette creation, which is modified from [5]. We propose a fast NER algorithm using large gazette and show that it significantly outperforms a naïve approach based on regular expressions. We describe experimental results obtained by using the system for gazette creation for various resume related named entities (e.g., ORG, DEGREE, EDUCATIONAL_INSTITUTE, DESIGNATION) and the associated NER on a large set of real-life resumes.