Classifying resources using a deep network

Patent metadata

Field	Value
Publication number	US-9449271-B2
Application number	US-201514834274-A
Country	US
Kind code	B2
Filing date	Aug 24, 2015
Priority date	Mar 13, 2013
Publication date	Sep 20, 2016
Grant date	Sep 20, 2016

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for scoring concept terms using a deep network. One of the methods includes receiving an input comprising a plurality of features of a resource, wherein each feature is a value of a respective attribute of the resource; processing each of the features using a respective embedding function to generate one or more numeric values; processing the numeric values using one or more neural network layers to generate an alternative representation of the features, wherein processing the floating point values comprises applying one or more non-linear transformations to the floating point values; and processing the alternative representation of the input using a classifier to generate a respective category score for each category in a pre-determined set of categories, wherein each of the respective category scores measure a predicted likelihood that the resource belongs to the corresponding category.

First claim

What is claimed is:

1. A method comprising:
receiving an input comprising a plurality of features of a resource, wherein each feature is a value of a respective attribute of the resource;
generating an alternative representation of the features of the resource, comprising:
generating a respective numeric representation of each of the features by processing each of the features using a respective embedding function, wherein each of the embedding functions is specific to features of a respective feature type, and
processing the respective numeric representations through one or more neural network layers to generate the alternative representation of the features of the resource; and

providing the alternative representation of the features of the resource as input to a neural network classifier for classification of the resource as belonging to one or more categories of a plurality of categories.

2. The method of claim 1 , wherein the neural network classifier is configured to:
process the alternative representation of the input to generate a respective category score for each of the plurality of categories, wherein each of the respective category scores measures a predicted likelihood that the resource belongs to the corresponding category.

3. The method of claim 2 , further comprising providing the category scores to a search system for use in determining whether or not index resources in a search engine index.

4. The method of claim 2 , further comprising providing the category scores to a search system for use in determining whether or not index resources in a search engine index.

5. The method of claim 2 , further comprising providing the category scores to a search system for use in generating and ordering search results in response to received search queries.

6. The method of claim 1 , wherein the numeric representations are vectors of floating point values.

7. The method of claim 1 , wherein the numeric representations are vectors of quantized integer values, and wherein an encoding of the quantized integer values represents floating point values.

8. The method of claim 1 , wherein the plurality of categories includes a search engine spam category.

9. The method of claim 1 , wherein the plurality of categories includes a respective category for each of a plurality of types of search engine spam.

10. The method of claim 1 , wherein the plurality of categories includes a respective category for each resource type in a group of resource types.

11. A system comprising one or more computers and one or more storage devices storing instructions that when executed by the one or more computers to perform operations comprising:
receiving an input comprising a plurality of features of a resource, wherein each feature is a value of a respective attribute of the resource;
generating an alternative representation of the features of the resource, comprising:
generating a respective numeric representation of each of the features by processing each of the features using a respective embedding function, wherein each of the embedding functions is specific to features of a respective feature type, and
processing the respective numeric representations through one or more neural network layers to generate the alternative representation of the features of the resource; and

providing the alternative representation of the features of the resource as input to a neural network classifier for classification of the resource as belonging to one or more categories of a plurality of categories.

12. The system of claim 11 , wherein the neural network classifier is configured to:
process the alternative representation of the input to generate a respective category score for each of the plurality of categories, wherein each of the respective category scores measures a predicted likelihood that the resource belongs to the corresponding category.

13. The system of claim 12 , the operations further comprising providing the category scores to a search system for use in determining whether or not index resources in a search engine index.

14. The system of claim 12 , the operations further comprising providing the category scores to a search system for use in determining whether or not index resources in a search engine index.

15. The system of claim 12 , the operations further comprising providing the category scores to a search system for use in generating and ordering search results in response to received search queries.

16. The system of claim 11 , wherein the plurality of categories includes a search engine spam category, and the category score for the search engine spam category measures a predicted likelihood that the resource is a search engine spam resource.

17. The system of claim 11 , wherein the plurality of categories includes a respective category for each of a plurality of types of search engine spam.

18. The system of claim 11 , wherein the plurality of categories includes a respective category for each resource type in a group of resource types.

19. One or more non-transitory storage media encoded with a computer program, the computer program comprising instructions that when executed by one or more computers to perform operations comprising:
receiving an input comprising a plurality of features of a resource, wherein each feature is a value of a respective attribute of the resource;
generating an alternative representation of the features of the resource, comprising:
generating a respective numeric representation of each of the features by processing each of the features using a respective embedding function, wherein each of the embedding functions is specific to features of a respective feature type, and
processing the respective numeric representations through one or more neural network layers to generate the alternative representation of the features of the resource; and

providing the alternative representation of the features of the resource as input to a neural network classifier for classification of the resource as belonging to one or more categories of a plurality of categories.

20. The computer storage media of claim 19 , wherein the neural network classifier is configured to:
process the alternative representation of the input to generate a respective category score for each of the plurality of categories, wherein each of the respective category scores measures a predicted likelihood that the resource belongs to the corresponding category.

Assignees

• Google Inc

Inventors

• Wang Qingzhou
• Liang Yu
• Yang Ke
• Chen Kai

Classifications

• G06N3/084 Primary

  Backpropagation, e.g. using gradient descent · CPC title

• G06F18/2413

  based on distances to training or reference patterns · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N3/042

  Knowledge-based neural networks; Logical representations of neural networks · CPC title

• G06N3/04 Primary

  Architecture, e.g. interconnection topology · CPC title