Abstract | ||
---|---|---|
We give a quantified reasoning and description of the perplexity for evaluating language models using the concept of entropy in information theory: The smaller the entropy of the language estimated by the language model is, the more precise the language model is; an interpolated model based on two (n-1)-gram models is better than the (n-1)-gram component models, but not a n-gram model. We also explore the methods to estimating the entropy of Chinese using language models. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1109/FSKD.2007.579 | FSKD (2) |
Keywords | Field | DocType |
gram model,quantified reasoning,gram component model,n-gram model,language models,language model,information analysis,information theory,entropy estimation,natural language processing,entropy,interpolated model,component model | Information theory,Entropy estimation,Perplexity,Computer science,Interpolation,Speech recognition,Information diagram,Natural language processing,Artificial intelligence,Machine learning,Language model | Conference |
Volume | ISBN | Citations |
2 | 978-0-7695-2874-8 | 0 |
PageRank | References | Authors |
0.34 | 4 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yangsen Zhang | 1 | 11 | 12.10 |
Gaijuan Huang | 2 | 0 | 1.01 |
Miao Mai | 3 | 0 | 0.34 |