Paper: Chinese Named Entity Recognition with Cascaded Hybrid Model

ACL ID N07-2050
Title Chinese Named Entity Recognition with Cascaded Hybrid Model
Venue Human Language Technologies
Session Short Paper
Year 2007
Authors
  • Xiaofeng Yu (The Chinese University of Hong Kong, Shatin Hong Kong)

We propose a high-performance cascaded hy- brid model for Chinese NER. Firstly, we use Boosting, a standard and theoretically well- foundedmachinelearningmethodtocombinea set of weak classifiers together into a base sys- tem. Secondly, we introduce various types of heuristic human knowledge into Markov Logic Networks (MLNs), an effective combination of first-order logic and probabilistic graphi- cal models to validate Boosting NER hypothe- ses. Experimental results show that the cas- caded hybrid model significantly outperforms the state-of-the-art Boosting model.