A decision tree method for finding and classifying names in Japanese texts

A supervised learning system which uses a decision tree to find and classify names in Japanese texts, which does not work fully automatically and also needs special dictionaries, but performs well and does not have parameters to be adjusted by hand.

Abstract

This paper describes a system whichusesadecision tree to ndandclassify names in Japanese texts. The decision tree uses part-of-speech, character type, and special dictionary information to determine the probability that aparticular type of name opens or closes at agiven position in the text. The output isgenerated from the consistent sequence of nameopens and name closes with the highest probability. This system does not require any human adjustment. Experiments indicate good accuracy with asmall amount of training data, and demonstrate the system's portability. Theissues of training data size and domain dependency are discussed. 1