other,0-1-C92-4199,bq great effectiveness of this approach . <term> Word Identification </term> has been an important and active
tech,10-1-C92-4199,bq been an important and active issue in <term> Chinese Natural Language Processing </term> . In this paper , a new mechanism
other,13-2-C92-4199,bq mechanism , based on the concept of <term> sublanguage </term> , is proposed for identifying <term>
other,19-2-C92-4199,bq </term> , is proposed for identifying <term> unknown words </term> , especially <term> personal names </term>
other,23-2-C92-4199,bq <term> unknown words </term> , especially <term> personal names </term> , in <term> Chinese newspapers </term>
other,27-2-C92-4199,bq especially <term> personal names </term> , in <term> Chinese newspapers </term> . The proposed mechanism includes
tech,4-3-C92-4199,bq </term> . The proposed mechanism includes <term> title-driven name recognition </term> , <term> adaptive dynamic word formation
tech,8-3-C92-4199,bq title-driven name recognition </term> , <term> adaptive dynamic word formation </term> , <term> identification of 2-character
tech,13-3-C92-4199,bq adaptive dynamic word formation </term> , <term> identification of 2-character and 3-character Chinese names without title </term> . We will show the experimental results
lr,8-4-C92-4199,bq show the experimental results for two <term> corpora </term> and compare them with the results
tool,17-4-C92-4199,bq compare them with the results by the <term> NTHU 's statistic-based system </term> , the only system that we know has
tech,9-5-C92-4199,bq significant improvements over the <term> WI systems </term> without the <term> name identification
other,13-5-C92-4199,bq <term> WI systems </term> without the <term> name identification </term> capability . This paper describes
hide detail