Generating Language Models

摘要

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating language models. In some implementations, data is accessed that indicates a set of classes corresponding to a concept. A first language model is generated in which a first class represents the concept. A second language model is generated in which second classes represent the concept. Output of the first language model and the second language model is obtained, and the outputs are evaluated. A class from the set of classes is selected based on evaluating the output of the first language model and the output of the second language model. In some implementations, the first class and the second class are selected from a parse tree or other data that indicates relationships among the classes in the set of classes.

主权项

1. A method performed by one or more computers, the method comprising:
accessing data indicating a set of classes corresponding to a concept, each of the classes representing a different set of words or phrases associated with the concept; generating a first language model in which a first class from the set of classes represents the concept; generating a second language model in which second classes from the set of classes represent the concept, each of the second classes being different from the first class; obtaining (i) output of the first language model for a set of input data, and (ii) output of the second language model for the set of input data; evaluating the output of the first language model and the output of the second language model; and selecting one or more classes from the set of classes based on evaluating the output of the first language model and the output of the second language model.