A Bayesian Interpretation of Interpolated Kneser-Ney
| dc.contributor.author | TEH, Yee Whye | en_US |
| dc.date.accessioned | 2006-02-02T09:23:28Z | en_US |
| dc.date.accessioned | 2017-01-23T07:00:02Z | |
| dc.date.available | 2006-02-02T09:23:28Z | en_US |
| dc.date.available | 2017-01-23T07:00:02Z | |
| dc.date.issued | 2006-02-02T09:23:28Z | en_US |
| dc.description.abstract | Interpolated Kneser-Ney is one of the best smoothing methods for n-gram language models. Previous explanations for its superiority have been based on intuitive and empirical justifications of specific properties of the method. We propose a novel interpretation of interpolated Kneser-Ney as approximate inference in a hierarchical Bayesian model consisting of Pitman-Yor processes. As opposed to past explanations, our interpretation can recover exactly the formulation of interpolated Kneser-Ney, and performs better than interpolated Kneser-Ney when a better inference procedure is used. | en_US |
| dc.format.extent | 1292400 bytes | en_US |
| dc.format.mimetype | application/pdf | en_US |
| dc.identifier.uri | https://dl.comp.nus.edu.sg/xmlui/handle/1900.100/1911 | en_US |
| dc.language.iso | en | en_US |
| dc.relation.ispartofseries | TRA2/06 | en_US |
| dc.title | A Bayesian Interpretation of Interpolated Kneser-Ney | en_US |
| dc.type | Technical Report | en_US |