Identifying relevant document languages through link context

摘要

Methods, systems, and apparatus, including computer program products, for identifying languages that are relevant to resource. In an aspect, language features are identified for incoming resource links to a resource and outgoing resource links from the resource. The language features or use by a language classification model to generate language relevance scores. The language relevance scores for each of the incoming resource links and outgoing resource links are used to generate a corresponding relevance measure for each of a plurality of languages. Each relevance measure is a measure of the relevance of the language to the resource.

主权项

1. A computer-implemented method, comprising:
selecting, by a data processing apparatus, a first resource having a corresponding first resource address and being linked to other resources by incoming resource links and outgoing resource links, each incoming resource link being a resource link that links to the first resource from a corresponding source resource, and each outgoing resource link being a resource link that links to a corresponding target resource from the first resource; generating, by the data processing apparatus, one or more language relevance scores for each of the incoming resource links and outgoing resource links from data defining the incoming resource links and the outgoing resource links, each of the language relevance scores representing a relevance of a corresponding language to the first resource; identifying, by the data processing apparatus, one or more languages as being relevant to the first resource from the language relevance scores; and associating with the resource in a resource index the one or more languages identified as being relevant to the first resource.