A maximum entropy approach to natural language processing
Computational Linguistics
Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Learn - Filter - Apply - Forget. Mixed Approaches to Named Entity Recognition
NLDB'01 Proceedings of the 6th International Workshop on Applications of Natural Language to Information Systems
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Shallow parsing with conditional random fields
NAACL '03 Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology - Volume 1
Automatic recognition of Chinese unknown words based on roles tagging
SIGHAN '02 Proceedings of the first SIGHAN workshop on Chinese language processing - Volume 18
Maximum entropy models for named entity recognition
CONLL '03 Proceedings of the seventh conference on Natural language learning at HLT-NAACL 2003 - Volume 4
Chinese named entity recognition based on multiple features
HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Personal Name Recognition Based on Categorized Linguistic Knowledge
WI-IAT '08 Proceedings of the 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology - Volume 03
A structural approach to extracting Chinese position relations from web pages
Journal of Web Engineering
Hi-index | 0.00 |
This paper presents the ability of Conditional Random Field (CRF) combining with multiple features to perform robust and accurate Chinese Named Entity Recognition. We describe the multiple feature templates including local feature templates and global feature templates used to extract multiple features with the help of human knowledge. Besides, we show that human knowledge can reasonably smooth the model and thus the need of training data for CRF might be reduced. From the experimental results on People's Daily corpus, we can conclude that our model is an effective pattern to combine statistical model and human knowledge. And the experiments on another data set also confirm the above conclusion, which shows that our features have consistence on different testing data.