Title :
Elastic neural networks for part of speech tagging
Author :
Ma, Qing ; Uchimoto, Kiyotaka ; Murata, Masaki ; Isahara, Hitoshi
Author_Institution :
Commun. Res. Lab., Minist. of Posts & Telecommun., Japan
Abstract :
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
Keywords :
grammars; linguistics; multilayer perceptrons; 3-layer perceptron; Thai; ambiguous words; elastic neural networks; part-of-speech tagging; statistical models; Frequency; Hidden Markov models; Information retrieval; Natural language processing; Neural networks; Optical character recognition software; Speech processing; Speech synthesis; Tagging; Training data;
Conference_Titel :
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5529-6
DOI :
10.1109/IJCNN.1999.835997
