Name
Title | ||
|---|---|---|
A Hybrid Method of Chinese Prosodic Word Tagging Based on Keyword Anchor and Hidden Markov Model |
Abstract | ||
|---|---|---|
In this paper, a new method of Chinese prosodic word tagging is presented. This method consists of a rule-based algorithm named “Keyword Anchor” and a statistical algorithm based on Hidden Markov Model (HMM). For keyword anchor algorithm, an anchor of the prosodic word is defined to help the system to find the whole prosodic word. For statistical algorithm, a length-based Hidden Markov Model (HMM) is used to find the best result of prosodic word tagging. The experiments of this method prove the better result than preceding methods in this field. The “Open Set F Score” of prosodic word based on this method is up to about 0.96. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/IALP.2009.24 | IALP |
Keywords | Field | DocType |
speech,labeling,data mining,read only memory,rule based,hmm,support vector machines,natural language processing,hidden markov models,hidden markov model | F1 score,Read-only memory,Pattern recognition,Computer science,Support vector machine,Statistical algorithm,Speech recognition,Artificial intelligence,Natural language processing,Hidden Markov model,Word processing,Open set | Conference |
Volume | Issue | ISSN |
null | null | 2159-1962 |
Citations | PageRank | References |
0 | 0.34 | 3 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Quan Zhou | 1 | 1 | 1.06 |
| Pan Deng | 2 | 0 | 0.34 |
| Hongjian Liu | 3 | 1 | 1.06 |
| Defeng Guo | 4 | 15 | 1.73 |
| Kenji Nagamatsu | 5 | 24 | 10.00 |