Automated resource selection process evaluation

• US 8,489,604 B1
• Filed: 10/26/2010
• Issued: 07/16/2013
• Est. Priority Date: 10/26/2010
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Sadovsky, Adam, Haahr, Paul, Strohman, Trevor, Xu, Jun, Shrauner, Jay, Bjornsson, Per, Schine, Gabriel
• Primary Examiner(s): Nguyen, Cindy
• Application Number:
  US12/912,229
• Time in Patent Office: 994 Days
• Field of Search: 707/706, 707/707, 707/711, 707/741, 707/737
• US Class Current: 707/737
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for evaluating resource selection processes. One method includes receiving test queries and generating a first group of resources corresponding to a first automated resource selection process and generating a second group of resources corresponding to a second automated resource selection process for each query. Another method includes generating a query results table for use in generating the groups of resources. The query results table maps queries to resources matched to the queries, and maps each resource to a respective score for the resource and the query, and one or more index selection signals for the resource.

Claims

1. A system comprising:

one or more computers configured to perform operations comprising;

receiving a plurality of test queries;

generating, for each test query, a first group of resources corresponding to a first automated resource selection process and a second group of resources corresponding to a second automated resource selection process, the generating comprising, for each test query;

identifying a plurality of resources responsive to the test query,

determining, for each resource in the plurality of resources, whether the first automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the first group of resources, and

determining, for each resource in the plurality of resources, whether the second automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the second group of resources,

wherein the determination whether the first automated resource selection process would classify a particular resource as to-be-indexed or not-to-be-indexed is made independently of the determination whether the second automated resource selection process would classify that particular resource as to-be-indexed or not-to-be-indexed.

2. The system of claim 1, wherein the determination whether the first automated resource selection process would classify any particular resource as to-be-indexed or not-to-be-indexed is made independently of the determination made for any other identified resource.

3. The system of claim 1, wherein the determination whether the second automated resource selection process would classify any particular resource as to-be-indexed or not-to-be-indexed is made independently of the determination made for any other identified resource.

4. The system of claim 1, wherein the operations further comprise determining, for at least one test query, that the first automated resource selection process would classify at least one resource as to-be-indexed and would classify at least one resource as not-to-be-indexed.

5. The system of claim 1, wherein the operations further comprise determining, for at least one test query, that the second automated resource selection process would classify at least one resource as to-be-indexed and would classify at least one resource as not-to-be-indexed.

6. The system of claim 1, wherein, for at least one of the test queries in the plurality of test queries, the first group of resources is different from the second group of resources.

7. The system of claim 1, wherein determining whether the first automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed comprises deriving an index selection score for the resource, determining that the first automated resource selection process would classify the resource as to-be-indexed if the index selection score satisfies a threshold, and otherwise determining that the first automated resource selection process would classify the resource as not-to-be-indexed.

8. The system of claim 1, wherein identifying a plurality of resources responsive to the test query comprises identifying the plurality of resources from a query results table, wherein the query results table maps each of a plurality of queries to a respective one or more resources responsive to the query and includes a query independent quality score for each resource.

9. The system of claim 1, wherein the operations further comprise:

receiving respective feedback for each test query, the respective feedback for each test query selecting either the first group of resources or the second group of resources; and

selecting either the first automated resource selection process or the second automated resource selection process as a result of the respective feedback for each test query.

10. The system of claim 9, wherein the operations further comprise presenting first search results corresponding to the first group of resources and second search results corresponding to the second group of resources to each evaluator in a plurality of evaluators, wherein:

receiving feedback selecting either the first group of resources or the second group of resources comprises receiving feedback from each evaluator indicating that the evaluator prefers the first search results or the second search results, and aggregating the received feedback.

11. The system of claim 10, wherein the operations further comprise obtaining a query-specific score for each resource in the first group of resources and obtaining a query-specific score for each resource in the second group of resources, wherein:

the first search results corresponding to the first group of resources have a first order derived from the query-specific score for each resource and presenting the first search results comprises presenting the first search results according to the first order; and

the second search results corresponding to the second group of resources have a second order derived from the query-specific score for each resource and presenting the second search results comprises presenting the second search results according to the second order.

12. The system of claim 1, wherein the operations further comprise:

for one or more of the test queries;

receiving the test query through a search engine user interface from each of a plurality of users;

presenting search results corresponding to the first group of resources to one or more first users in the plurality of users in response to receiving the test query from each first user, and presenting search results corresponding to the second group of resources to one or more second users in the plurality of users in response to receiving the test query from each second user; and

comparing user assessment of the search results corresponding to the first group of resources and user assessment of search results corresponding to the second group of resources, resulting in a comparison; and

analyzing the comparison for each of the one or more test queries to select either the first automated resource selection process or the second automated resource selection process.

13. The system of claim 12, wherein the operations further comprise obtaining a query-specific score for each resource in the first group of resources and obtaining a query-specific score for each resource in the second group of resources, wherein:

the search results for a first subset of identified resources have a first order derived from the query-specific scores for each resource and presenting the search results comprises presenting the search results according to the first order; and

the search results for a second subset of identified resources have a second order derived from the query-specific scores for each resource and presenting the search results comprises presenting the search results according to the second order.

14. A computer-implemented method, comprising:

receiving a plurality of test queries;

generating, for each test query, a first group of resources corresponding to a first automated resource selection process and a second group of resources corresponding to a second automated resource selection process, the generating comprising, for each test query;

identifying a plurality of resources responsive to the test query,

determining, for each resource in the plurality of resources, whether the first automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the first group of resources, and

determining, for each resource in the plurality of resources, whether the second automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the second group of resources,

wherein the determination whether the first automated resource selection process would classify a particular resource as to-be-indexed or not-to-be-indexed is made independently of the determination whether the second automated resource selection process would classify that particular resource as to-be-indexed or not-to-be-indexed.

15. The method of claim 14, wherein the determination whether the first automated resource selection process would classify any particular resource as to-be-indexed or not-to-be-indexed is made independently of the determination made for any other identified resource.

16. The method of claim 14, wherein the determination whether the second automated resource selection process would classify any particular resource as to-be-indexed or not-to-be-indexed is made independently of the determination made for any other identified resource.

17. The method of claim 14, further comprising:

determining, for at least one test query, that the first automated resource selection process would classify at least one resource as to-be-indexed and would classify at least one resource as not-to-be-indexed.

18. The method of claim 14, further comprising:

determining, for at least one test query, that the second automated resource selection process would classify at least one resource as to-be-indexed and would classify at least one resource as not-to-be-indexed.

19. The method of claim 14, wherein, for at least one of the test queries in the plurality of test queries, the first group of resources is different from the second group of resources.

20. The method of claim 14, wherein determining whether the first automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed comprises deriving an index selection score for the resource, determining that the first automated resource selection process would classify the resource as to be indexed if the index selection score satisfies a threshold, and otherwise determining that the first automated resource selection process would classify the resource as not-to-be-indexed.

21. The method of claim 14, wherein identifying a plurality of resources responsive to the test query comprises identifying the plurality of resources from a query results table, wherein the query results table maps each of a plurality of queries to a respective one or more resources responsive to the query and includes a query independent quality score for each resource.

22. The method of claim 14, further comprising:

receiving respective feedback for each test query, the respective feedback for each test query selecting either the first group of resources or the second group of resources; and

selecting either the first automated resource selection process or the second automated resource selection process as a result of the respective feedback for each test query.

23. The method of claim 22, further comprising presenting first search results corresponding to the first group of resources and second search results corresponding to the second group of resources to each evaluator in a plurality of evaluators, wherein:

receiving feedback selecting either the first group of resources or the second group of resources comprises receiving feedback from each evaluator indicating that the evaluator prefers the first search results or the second search results, and aggregating the received feedback.

24. The method of claim 23, further comprising obtaining a query specific score for each resource in the first group of resources and obtaining a query specific score for each resource in the second group of resources, wherein:

the first search results corresponding to the first group of resources have a first order derived from the query specific score for each resource and presenting the first search results comprises presenting the first search results according to the first order; and

the second search results corresponding to the second group of resources have a second order derived from the query specific score for each resource and presenting the second search results comprises presenting the second search results according to the second order.

25. The method of claim 14, further comprising:

for one or more of the test queries;

receiving the test query through a search engine user interface from each of a plurality of users;

presenting search results corresponding to the first group of resources to one or more first users in the plurality of users in response to receiving the test query from each first user, and presenting search results corresponding to the second group of resources to one or more second users in the plurality of users in response to receiving the test query from each second user; and

comparing user assessment of the search results corresponding to the first group of resources and user assessment of search results corresponding to the second group of resources, resulting in a comparison; and

analyzing the comparison for each of the one or more test queries to select either the first automated resource selection process or the second automated resource selection process.

26. The method of claim 25, further comprising obtaining a query specific score for each resource in the first group of resources and obtaining a query specific score for each resource in the second group of resources, wherein:

the search results for the a first subset of identified resources have a first order derived from the query specific scores for each resource and presenting the search results comprises presenting the search results according to the first order; and

the search results for the a second subset of identified resources have a second order derived from the query specific scores for each resource and presenting the search results comprises presenting the search results according to the second order.

27. A computer storage medium encoded with a computer program, the program comprising instructions that when executed by data processing apparatus cause the data processing apparatus to perform operations comprising:

receiving a plurality of test queries;

generating, for each test query, a first group of resources corresponding to a first automated resource selection process and a second group of resources corresponding to a second automated resource selection process, the generating comprising, for each test query;

identifying a plurality of resources responsive to the test query,

determining, for each resource in the plurality of resources, whether the first automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the first group of resources, and

determining, for each resource in the plurality of resources, whether the second automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the second group of resources,

wherein the determination whether the first automated resource selection process would classify a particular resource as to-be-indexed or not-to-be-indexed is made independently of the determination whether the second automated resource selection process would classify that particular resource as to-be-indexed or not-to-be-indexed.

28. A system comprising:

one or more computers configured to perform operations comprising;

receiving a plurality of test queries;

selecting, for each test query, a first group of resources corresponding to a first automated resource selection process and a second group of resources corresponding to a second automated resource selection process, the selecting comprising, for each test query;

identifying a plurality of resources responsive to the test query,

determining, for each resource in the plurality of resources, whether the first automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the first group of resources, and

determining, for each resource in the plurality of resources, whether the second automated resource selection process would classify the resource as to-be-indexed or not-to-be-indexed, and then selecting all resources classified as to-be-indexed as the second group of resources;

obtaining a respective query-specific score for each resource in the first group of resources and each resource in the second group of resources;

presenting for comparison by a user first search results for a first subset of the first group of resources and second search results for a second subset of the second group of resources, wherein the first search results are presented in an order derived from the respective query-specific scores for the resources in the first subset and the second search results are presented in an order derived from the respective query-specific scores for the resources in the second subset.

29. The system of claim 28, wherein the operations further comprise:

comparing an assessment of the first search results by the user and an assessment of the second search results by the user; and

selecting either the first automated search process or the second automated search process based on a result of the comparing.

30. A method comprising:

receiving a plurality of test queries;

selecting, for each test query, a first group of resources corresponding to a first automated resource selection process and a second group of resources corresponding to a second automated resource selection process, the selecting comprising, for each test query;

identifying a plurality of resources responsive to the test query,

determining, for each resource in the plurality of resources, whether the first automated resource selection process would classify the resource as to be indexed or not to be indexed, and then selecting all resources classified as to be indexed as the first group of resources, and

determining, for each resource in the plurality of resources, whether the second automated resource selection process would classify the resource as to be indexed or not to be indexed, and then selecting all resources classified as to be indexed as the second group of resources;

obtaining a respective query specific score for each resource in the first group of resources and each resource in the second group of resources;

presenting for comparison by a user first search results for a first subset of the first group of resources and second search results for a second subset of the second group of resources, wherein the first search results are presented in an order derived from the respective query specific scores for the resources in the first subset and the second search results are presented in an order derived from the respective query specific scores for the resources in the second subset.

31. The method of claim 30, further comprising:

comparing an assessment of the first search results by the user and an assessment of the second search results by the user; and

selecting either the first automated search process or the second automated search process based on a result of the comparing.