| 摘要 |
In order to solve a conventional problem that an accurate translation cannot be realized, a learning apparatus includes: a parser unit that parses a target language sentence contained in one or more pairs of original and translated sentences, thereby acquiring a binary tree of the target language sentence having one or more target language partial structures; a source language element acquiring unit that acquires one or more source language elements; a source language partial structure acquiring unit that acquires one or more source language partial structures each containing a parent node having a phrase label and two child nodes each having a phrase label or a source language element, using a structure indicated by the one or more target language partial structures contained in the binary tree of the target language sentence; a labeling unit that provides a reordering label to the one or more source language partial structures; a model building unit that builds one or more parsing models each having appearance probability information regarding a labeled source language partial structure, using one or more labeled source language partial structures; and an accumulating unit that accumulates a binary tree of a source language sentence having the one or more parsing models. Accordingly, it is possible to learn a model that can realize an accurate translation. |
