Placement of search results using user intent

• US 8,650,173 B2
• Filed: 06/23/2010
• Issued: 02/11/2014
• Est. Priority Date: 06/23/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: McDonald, Kieran Richard, Gilad-Bachrach, Ran, Malhotra, Nipoon, Agrawal, Nitin, Ahari, Sanaz
• Primary Examiner(s): VO, TRUONG V
• Application Number:
  US12/821,134
• Publication Number:
  US 20110320440A1
• Time in Patent Office: 1,329 Days
• Field of Search: 707/706, 707/712
• US Class Current: 707/706
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking

Abstract

Technologies are described herein for placing search results on a search engine results page (SERP). A query may be received. The query may be transmitted to a plurality of search result providers. A first set of search results and a second set of search results may be received from the search result providers. Intent features may be extracted from the first set of search results. User intent of the second set of search results may be inferred based on the extracted intent features. The first set of search results and the second set of search results may be ranked based on the inferred user intent. The SERP may be rendered according to the ranked first set and second set of search results.

Claims

1. A computer-implemented method for placing search results on a search engine results page (SERP), the method comprising computer-implemented operations for:

receiving, through a computer having a memory and a processor, a single query;

transmitting, through the computer, the query to a plurality of search result providers;

upon transmitting the query to the plurality of search result providers, receiving, through the computer, a first set of search results and a second set of search results;

extracting, through the computer, intent features from the first set of search results by extracting correlations and anti-correlations between the first set of search results and user related information;

inferring, through the computer, user intent of the second set of search results based on the extracted intent features;

ranking, through the computer, the first set of search results and the second set of search results based on the inferred user intent; and

rendering, through the computer, the SERP according to the ranked first set and second set of search results.

2. The computer-implemented method of claim 1, wherein extracting, through the computer, intent features from the first set of search results comprises extracting, through the computer, text contained in the first set of search results.

3. The computer-implemented method of claim 2, wherein extracting, through the computer, text contained in the first set of search results comprises extracting, through the computer, single words from the first set of search results.

4. The computer-implemented method of claim 2, wherein extracting, through the computer, text contained in the first set of search results comprises extracting, through the computer, phrases containing two or more words from the first set of search results.

5. The computer-implemented method of claim 2, wherein the text contained in the first set of search results comprises at text from uniform resource locators (URLs), titles, or snippets in the first set of search results.

6. The computer-implemented method of claim 1, wherein extracting, through the computer, intent features from the first set of search results further comprises extracting, through the computer, similarities and differences between the first set of search results from different search result providers.

7. The computer-implemented method of claim 1, wherein extracting, through the computer, intent features from the first set of search results further comprises extracting, through the computer, information known about websites corresponding to uniform resource locators (URLs) in the first set of search results.

8. The computer-implemented method of claim 1, wherein the first set of search results comprise search results having one or more information types;

and wherein the second set of search results comprise search results having one or more other information types.

9. The computer-implemented method of claim 1, wherein inferring, through the computer, user intent based on the extracted intent features comprises determining, through the computer, an intent value for each of one or more information types based on the intent features, each of the intent values indicating a relevance of the corresponding information type to the query.

10. The computer-implemented method of claim 9, wherein each of the intent values comprises a binary value indicating whether the corresponding information type is relevant to the query.

11. The computer-implemented method of claim 9, wherein each of the intent values comprises a range indicating a level of relevance between the corresponding information type and the query.

12. The computer-implemented method of claim 9, wherein ranking, through the computer, the first set of search results and the second set of search results based on the inferred user intent comprises:

assigning, through the computer, a higher rank for search results in the second set of search results that are relevant to the query according to the intent value; and

assigning, through the computer, a lower rank for search results in the second set of search results that are not relevant to the query according to the intent value.

13. The computer-implemented method of claim 1 and further comprising:

generating a classification model through a machine learning technology; and

wherein inferring user intent based on the extracted intent features comprises inferring user intent based on the extracted intent features according to the classification model.

14. A computer system, comprising:

a processor;

a memory communicatively coupled to the processor; and

a search application (i) which executes in the processor from the memory and (ii) which, when executed by the processor, causes the computer system to place search results on a search engine results page (SERP) by

receiving a single query,

transmitting the query to a plurality of search result providers,

upon transmitting the query to the plurality of search result providers, receiving a first set of search results having one or more information types and a second set of search results having one or more other information types,

extracting intent features from the first set of search results,

inferring user intent of the second set of search results based on the extracted intent features,

ranking the first set of search results and the second set of search results based on the inferred user intent by validating correlations and anti-correlations between features in the first set of search results and the second set of search results, and

rendering the SERP according to the ranked first set and second set of search results.

15. The computer system of claim 14, wherein the information types include web page results, map results, news results, video results, image results, stock quote results, commerce results, or local search results.

16. The computer system of claim 14, wherein the search application, when executed by the processor, further causes the computer system to place search results on the SERP by generating a classification model through a machine learning technology;

and wherein inferring user intent based on the extracted intent features comprises inferring user intent based on the extracted intent features according to the classification model.

17. A computer-readable storage medium having computer-executable instructions stored thereon which, when executed by a computer, cause the computer to:

receive a single query;

transmit the query to a plurality of search result providers;

upon transmitting the query to the plurality of search result providers, receive a first set of search results having one or more information types and a second set of search results having one or more other information types, the first set of search results comprising web page results, the second set of search results comprising map results, news results, video results, and image results;

extract intent features from the web page results by extracting correlations and anti-correlations between the web page results and user related information;

infer user intent of the web page results, the map results, the news results, the video results, and the image results based on the extracted intent features;

rank the web page results, the map results, the news results, the video results, and the image results based on the inferred user intent; and

render the search engine results page according to the ranked web pages results, map results, news results, video results, and image results.