Propagating query classifications

• US 8,838,587 B1
• Filed: 04/19/2010
• Issued: 09/16/2014
• Est. Priority Date: 04/19/2010
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Adams, Henele I., Kim, Hyung-Jin
• Primary Examiner(s): Bhatia, Ajay
• Assistant Examiner(s): HUANG, MIRANDA M
• Application Number:
  US12/762,787
• Time in Patent Office: 1,611 Days
• Field of Search: None
• US Class Current: 707/731
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

In general, one aspect described can be embodied in a method for determining a classification for a query. The method can include receiving a request to determine whether to assign a classification to a first query, identifying a plurality of search entities that are associated with the first query based upon data associated with each of the plurality of search entities and the first query, and determining whether to assign the classification to the first query based upon classifications for the identified search entities.

Claims

1. A computer-implemented method comprising:

receiving, at a computer system, a request to determine whether to assign a classification to a first query;

selecting, by the computer system, a plurality of search entities that are associated with the first query based on respective user behavior data for the first query associated with each of the search entities;

determining a first measure of how many of the plurality of search entities have been assigned the classification;

determining that the first measure of how many of the plurality of search entities have been assigned the classification satisfies a classification threshold;

in response to determining that the first measure of how many of the plurality of search entities have been assigned the classification satisfies the classification threshold, determining a second measure of how many of a top group of search entities from the plurality of search entities have been assigned the classification, each search entity in the top group of search entities being in a subset of the plurality of search entities having highest respective measures of relevance to the query based on the user behavior data;

determining, by the computer system that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification; and

in response to determining that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification, assigning the classification to the first query.

2. The method of claim 1, wherein the plurality of identified search entities comprises a plurality of documents, and wherein the respective user behavior data associated with each document of the plurality of documents represents user behavior relative to the document when provided as a search result for the first query.

3. The method of claim 1, wherein the plurality of identified search entities comprises a plurality of sites, and wherein the respective user behavior data associated with each site of the plurality of sites represents user behavior relative to one or more documents on the site when the one or more documents were provided as search results for the first query.

4. The method of claim 1, wherein the plurality of identified search entities comprises a plurality of web domains, and wherein the respective user behavior data associated with each web domain of the plurality of web domains represents user behavior relative to one or more documents in the web domain when the one or more documents were provided as search results for the first query.

5. The method of claim 1, wherein the plurality of identified search entities comprises a plurality of second queries distinct from the first query.

6. The method of claim 5, wherein the respective user behavior data associated with each second query of the plurality of second queries represents a likelihood of the first query being submitted in a same query session as the second query.

7. The method of claim 1, further comprising:

selecting a subgroup of search entities from the top group of search entities, each search entity in the subgroup of search entities being in a subset of the top group of search entities having highest respective measures of relevance to the query based on the user behavior data; and

determining a third measure of how many of the subgroup of search entities have been assigned the classification,

wherein assigning the classification to the first query is further based on determining whether the third measure is consistent with the second measure, wherein the third measure is determined to be consistent with the second measure when the third measure is within a consistency threshold of the second measure.

8. The method of claim 1, wherein determining the first measure of how many of the plurality of search entities have been assigned the classification comprises:

determining the first measure of how many of the plurality of search entities have been assigned the classification according to respective weights for the classification assigned to each of the plurality of search entities, the weights being based on the user behavior data associated with each of the plurality of search entities.

9. The method of claim 1, wherein selecting the plurality of search entities that are associated with the first query based on respective user behavior data for the first query associated with each of the search entities comprises selecting search entities having at least a threshold amount of user behavior data associated with the first query.

10. The method of claim 1, further comprising providing the classification assigned to the first query as input to a document ranking process for ranking a plurality of documents in response to the first query.

11. The method of claim 1, wherein a value of the classification threshold is based on a type of the classification.

21. The method of claim 1, wherein determining that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification comprises determining that the second measure satisfies the classification threshold or determining that the second measure is within a consistency threshold of the first measure.

12. A computer program product encoded on a non-transitory computer-readable medium, operable to cause a data processing apparatus to perform operations comprising:

receiving a request to determine whether to assign a classification to a first query;

selecting a plurality of search entities that are associated with the first query based on respective user behavior data for the first query associated with each of the search entities;

determining a first measure of how many of the plurality of search entities have been assigned the classification;

determining that the first measure of how many of the plurality of search entities have been assigned the classification satisfies a classification threshold;

in response to determining that the first measure of how many of the plurality of search entities have been assigned the classification satisfies the classification threshold, determining a second measure of how many of a top group of search entities from the plurality of search entities have been assigned the classification, each search entity in the top group of search entities being in a subset of the plurality of search entities having highest respective measures of relevance to the query based on the user behavior data;

determining that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification; and

in response to determining that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification, assigning the classification to the first query.

13. The computer program product of claim 12, wherein the plurality of identified search entities comprises a plurality of documents, and wherein the respective user behavior data associated with each document of the plurality of documents represents user behavior relative to the document when provided as a search result for the first query.

14. The computer program product of claim 12, wherein the plurality of identified search entities comprises a plurality of sites, and wherein the respective user behavior data associated with each site of the plurality of sites represents user behavior relative to one or more documents on the site when the one or more documents were provided as search results for the first query.

15. The computer program product of claim 12, wherein the plurality of identified search entities comprises a plurality of web domains, and wherein the respective user behavior data associated with each web domain of the plurality of web domains represents user behavior relative to one or more documents in the web domain when the one or more documents were provided as search results for the first query.

16. The computer program product of claim 12 wherein the plurality of identified search entities comprises a plurality of second queries distinct from the first query.

17. The computer program product of claim 16, wherein the respective user behavior data associated with each second query of the plurality of second queries represents a likelihood of the first query being submitted in a same query session as the second query.

22. The computer program product of claim 12, wherein determining that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification comprises determining that the second measure satisfies the classification threshold or determining that the second measure is within a consistency threshold of the first measure.

18. A system for determining a classification for a query, the system comprising:

one or more servers;

an interface to the one or more servers configured to receive a request to determine whether to assign a classification to a first query;

a search entity classifier engine installed on the one or more servers that is configured to select a plurality of search entities that are associated with the first query based on respective user behavior data for the first query associated with each of the search entities; and

a query classifier engine installed on the one or more servers that is configured to;

determine a first measure of how many of the plurality of search that have been assigned the classification;

determine that the first measure of how many of the plurality of search entities have been assigned the classification satisfies a classification threshold;

in response to determining that the first measure of how many of the plurality of search entities have been assigned the classification satisfies the classification threshold, determine a second measure of how many of a top group of search entities from the plurality of search entities have been assigned the classification, each search entity in the top group of search entities being in a subset of the plurality of search entities having highest respective measures of relevance to the query based on the user behavior data;

determine that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification; and

in response to determining that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification, assign the classification to the first query.

19. The system of claim 18, wherein the plurality of identified search entities comprises a plurality of documents, and wherein the respective user behavior data associated with each document of the plurality of documents represents user behavior relative to the document when provided as a search result for the first query.

20. The system of claim 18, wherein the plurality of identified search entities comprises a plurality of sites, and wherein the respective user behavior data associated with each site of the plurality of sites represents user behavior relative to one or more documents on the site when the one or more documents were provided as search results for the first query.

23. The system of claim 18, wherein determining that the first measure of how many of the plurality of search entities have been assigned the classification is consistent with the second measure of how many of the top group of search entities have been assigned the classification comprises determining that the second measure satisfies the classification threshold or determining that the second measure is within a consistency threshold of the first measure.