A Two-Stage Bi-LSTM Model for Chinese Company Name Recognition

Chinese company name is a special name entity of organization, which plays a critical role in multiple application scenarios. Traditional rule-based or statistic based approaches that can achieve effective recognition for a company name at restriction environment, which is tricky to tailor the demands of real application scenarios. In this paper, we propose a two-stage Bi-LSTM model to achieve the Chinese company name recognition. The first stage is to detect the candidate Chinese company name by Bi-LSTM-CRF Model, and then the second stage is to further identify the company name via Bi-LSTM. We conduct the comparison experiment on a labelled benchmark dataset, the proposed approach achieves the 98.8% precision, 83.7% recall rate and 90.62% F-measure. It significantly outperforms traditional rule-based and statistics based approaches.

[1]  Nitish Srivastava,et al.  Improving neural networks by preventing co-adaptation of feature detectors , 2012, ArXiv.

[2]  Mónica Marrero,et al.  Named Entity Recognition: Fallacies, challenges and opportunities , 2013, Comput. Stand. Interfaces.

[3]  Ralph Grishman,et al.  Message Understanding Conference- 6: A Brief History , 1996, COLING.

[4]  L. F. Rau,et al.  Extracting company names from text , 1991, [1991] Proceedings. The Seventh IEEE Conference on Artificial Intelligence Application.

[5]  Lei Meng,et al.  An Improved Method for Chinese Company Name and Abbreviation Recognition , 2017, KMO.

[6]  Wei Li,et al.  Early results for Named Entity Recognition with Conditional Random Fields, Feature Induction and Web-Enhanced Lexicons , 2003, CoNLL.

[7]  Scharolta Katharina Siencnik.  Adapting word2vec to Named Entity Recognition , 2015, NODALIDA.

[8]  Jian Su,et al.  Named Entity Recognition using an HMM-based Chunk Tagger , 2002, ACL.

[9]  Thorsten Joachims,et al.  Text Categorization with Support Vector Machines: Learning with Many Relevant Features , 1998, ECML.

[10]  Quoc V. Le,et al.  Distributed Representations of Sentences and Documents , 2014, ICML.

[11]  Andrew McCallum,et al.  A comparison of event models for naive bayes text classification , 1998, AAAI 1998.

[12]  Nanyun Peng,et al.  Improving Named Entity Recognition for Chinese Social Media with Word Segmentation Representation Learning , 2016, ACL.

[13]  Xiang Zhang,et al.  Character-level Convolutional Networks for Text Classification , 2015, NIPS.

[14]  Eric Nichols,et al.  Named Entity Recognition with Bidirectional LSTM-CNNs , 2015, TACL.

[15]  Wanxiang Che,et al.  Named Entity Recognition with Bilingual Constraints , 2013, HLT-NAACL.

[16]  Eduard H. Hovy,et al.  End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF , 2016, ACL.

[17]  Zhenchao Jiang,et al.  Biomedical named entity recognition based on extended Recurrent Neural Networks , 2015, 2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[18]  Sebastian Thrun,et al.  Text Classification from Labeled and Unlabeled Documents using EM , 2000, Machine Learning.

[19]  Franck Dernoncourt,et al.  Sequential Short-Text Classification with Recurrent and Convolutional Neural Networks , 2016, NAACL.

[20]  Soumya K. Ghosh,et al.  Conditional Random Field Based Named Entity Recognition in Geological Text , 2010 .

[21]  Burr Settles,et al.  Biomedical Named Entity Recognition using Conditional Random Fields and Rich Feature Sets , 2004, NLPBA/BioNLP.

[22]  Olivier Galibert,et al.  Named and Specific Entity Detection in Varied Data: The Quæro Named Entity Baseline Evaluation , 2010, LREC.