Elastic neural networks for part of speech tagging
Qing Ma, Kiyotaka Uchimoto, Masaki Murata, Hitoshi Isahara · 2003
This paper presents a part of speech (POS) neuro tagger which consists of a 3-layer perceptron with elastic input. Computer experiments show that the neuro tagger has an accuracy of 94.4% for tagging ambiguous words when a small Thai corpus with 22,311 ambiguous words is used for training. A series of comparative experiments further show that the neuro tagger is definitely far superior to the statistical models including the frequency model (a base-line model), local n-gram model, and HMM.