Re-ranking top search results

• US 8,661,030 B2
• Filed: 04/09/2009
• Issued: 02/25/2014
• Est. Priority Date: 04/09/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: El-Saban, Motaz Ahmad, Burges, Christopher John Champness, Wu, Qiang
• Primary Examiner(s): Corrielus, Jean M
• Application Number:
  US12/421,022
• Publication Number:
  US 20100262612A1
• Time in Patent Office: 1,783 Days
• Field of Search: 707/752, 707/723, 707/760, 715/753
• US Class Current: 707/723
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

The claimed subject matter provides a system and/or a method that facilitates generating sorted search results for a query. An interface component can receive a query in a first language. A first ranker can be trained from a portion of data related to a second language. A second ranker can correspond to the first language, wherein the second ranker is untrained due to a limited amount of data related to the first language. A sorting component can invoke the first ranker to generate and order a pre-defined number of search results for the received query and subsequently invoke the second ranker to the pre-defined number of search results to generate a re-ordered number of search results in the first language for the received query.

Claims

1. A computer-implemented method that facilitates generating a search result for a received query, comprising:

receiving a query in a first human language;

evaluating a first language ranker that corresponds to the first human language to ascertain if the first language ranker satisfies one or more criteria to be classified as a trained ranker that is associated with a particular degree of training;

upon ascertaining that the first language ranker does not satisfy the one or more criteria, selecting a second language ranker that corresponds to a second human language that is disparate from the first human language;

upon selecting the second language ranker, employing the second language ranker to generate a set of search results for the received query; and

employing the first language ranker to generate a re-ordering of the set of search results.

2. The method of claim 1, wherein the second language ranker is trained by a greater amount of training data than the first language ranker.

3. A system comprising:

one or more processors;

memory communicatively coupled to the one or more processors;

an interface component maintained in the memory and executable by the one or more processors to receive a query in a first human language;

a first ranker maintained in the memory and executable by the one or more processors to generate ordered search results for the query, the first ranker being trained in a second human language; and

a second ranker maintained in the memory and executable by the one or more processors to generate reordered search results from the ordered search results, the second ranker being trained in the first human language and being trained to a lesser degree in the first human language than the first ranker is trained in the second human language.

4. The system of claim 3, wherein at least one of the first ranker or the second ranker is trained based at least in part on input from a user that indicates a degree of relevance a search result has to a query.

5. The system of claim 3, wherein the second ranker is trained to a lesser degree than the first ranker due to an amount of training data that is used to train the second ranker being less than an amount of training data that is used to train the first ranker.

6. The system of claim 3, wherein the second ranker is configured to generate reordered search results by:

selecting a predefined number of top results of the ordered search results that were generated by the first ranker; and

reordering the selected results.

7. The system of claim 3, wherein at least one of the first ranker or the second ranker is configured to utilize a search engine to provide one or more search results.

8. The system of claim 3, further comprising a query evaluator maintained in the memory and executable by the one or more processors to identify the first language of the query based at least in part on an originating Internet Protocol (IP) address associated with the query.

9. The system of claim 3, wherein the query comprises at least one of a query in a written form or a query in a spoken form.

10. The system of claim 3, further comprising a ranker selector maintained in the memory and executable by the one or more processors to select at least one of the first ranker or the second ranker based at least in part on a language of the query.

11. The system of claim 10, wherein the ranker selector is configured to select the first ranker to generate the ordered search results based at least in part on an amount of training data that was used to train the first ranker.

12. The system of claim 3, wherein at least one of the first ranker or the second ranker is configured to utilize a feature vector to generate a relevancy score to provide at least one search result.

13. The system of claim 3, wherein the query relates to a data search associated with at least one of the Internet, a data store, a network, a website, or a server.

14. One or more computer storage media storing computer-readable instructions that, when executed, instruct one or more processors to perform acts comprising:

receiving a query in a first human language;

determining that a first ranker that is trained for the first human language does not satisfy one or more criteria to be classified as a trained ranker that is associated with a particular degree of training;

upon determining that the first ranker does not satisfy the one or more criteria, selecting a second ranker that is trained for a second human language, the second ranker being trained with a greater amount of training data than the first ranker;

utilizing the second ranker to generate a set of search results for the query;

utilizing the first ranker to reorder a predefined number of search results of the set of search results; and

providing the reordered search results.

15. The one or more computer storage media of claim 14, wherein the predefined number of search results of the set of search results comprises a number of top search results of the set of search results.

16. The one or more computer storage media of claim 14, wherein at least one of the first ranker or the second ranker utilizes a feature vector to generate a relevancy score to provide at least one search result.

17. The one or more computer storage media of claim 14, wherein the query relates to a data search associated with at least one of the Internet, a data store, a network, a website, or a server.

18. The one or more computer storage media of claim 14, wherein at least one of the first ranker or the second ranker is trained based at least in part on input from a user that indicates a degree of relevance a search result has to a query.

19. The one or more computer storage media of claim 14, wherein the second ranker is selected based at least in part on an amount of training data that was used to train the second ranker.