Freshness-based ranking

• US 8,832,088 B1
• Filed: 07/30/2012
• Issued: 09/09/2014
• Est. Priority Date: 07/30/2012
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Chen, Zhihui, Zhu, Cheng
• Primary Examiner(s): Bhatia, Ajay
• Assistant Examiner(s): Le, Jessica N
• Application Number:
  US13/562,098
• Time in Patent Office: 771 Days
• Field of Search: 707/705, 707/723
• US Class Current: 707/723
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• 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
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for ranking search results. One of the methods includes receiving a search result obtained in response to a query. A recent impression probability for the query for a recent time period and an overall impression probability for the query for an overall time period are computed. A ratio QtoA of the recent impression probability to the overall impression probability is computed. A determination is made that users prefer newer resources over older resources for the query based on QtoA and that the resource is a new resource. A new score S′ is associated with the resource in place of S, wherein the new score S′ signifies a better result than the score S signifies, based on determining that users prefer newer resources over older resources for the query, and that the resource is a new resource.

Claims

1. A computer-implemented method comprising:

receiving a search result obtained in response to a query, wherein the search result identifies a resource and has an associated score S;

computing a recent impression probability for the query for a recent time period and an overall impression probability for the query for an overall time period,

wherein each impression probability corresponds to a ratio of (i) a count of search result impressions selected by users to (ii) a count of all search result impressions presented to users in the respective time periods,

wherein the search result impressions were impressions provided in response to the query by a search engine during the respective time period, and

wherein the overall time period is a time period that began before the recent time period and is longer than the recent time period;

computing a QtoA ratio of the recent impression probability to the overall impression probability;

determining that users prefer newer resources over older resources for the query based on the QtoA ratio;

determining that the resource is a new resource; and

associating a new score S′

with the resource in place of S, wherein the new score S′

signifies a better result than the score S signifies, based on determining, that users prefer newer resources over older resources for the query and that the resource is a new resource.

2. The method of claim 1, wherein the overall time period includes the recent time period.

3. The method of claim 1, wherein the recent time period includes at least one previous day, month, or year.

4. The method of claim 1, wherein a search result impression is a search result that is selected by a user or ranked above a search result that is selected by a user.

5. The method of claim 1, wherein S′

has a boosted value that is higher relative to boosted scores of other resources that are older than the resource.

6. The method of claim 1, further comprising:

determining a query value Q for the query, the query value being based on the QtoA ratio;

determining a resource value D for the resource, the resource value being based on a function of an age of the resource, F(age), wherein age is the age of the resource; and

calculating S′

as a product of S and a value of a boosting function of the query value Q and the resource value D.

7. The method of claim 6, wherein the boosting function is a continuous function having a range of values that includes values less than 1, a value equal to 1, and values greater than 1, whereby the product of S and the value of the boosting function can be less than S, equal to S, or greater than S.

8. The method of claim 6, wherein the boosting function has the form f (Q, D)=Q^D.

9. The method of claim 6, wherein determining Q comprises determining Q from a maximum Max(QFval) of one or more of the following query freshness values:

a percentile value of a number of occurrences of the query being received within a first recent time period;

percentile value of a number of occurrences of the query on blog web pages within a second recent time period;

a percentile value of a number of occurrences of the query on news web pages within a third recent time period;

a percentile value of a number of occurrences of the query on social network web pages within a fourth recent time period;

a percentile value of a number of occurrences of the query requesting news search results within a fifth recent time period;

a percentile value of a ratio of a number of occurrences of the query requesting news search results within a sixth recent time period to a number of occurrences of the query requesting web search results within the sixth recent time period;

a percentile value of a number of user selections of news search results provided in response to the query within a seventh recent time period; and

a percentile value of a ratio of a number of user selections of news search results provided in response to the query within an eighth recent time period to a number of user selections of web search results provided in response to the query within the eighth recent time period.

10. The method of claim 9, wherein Q satisfies:

11. The method of claim 9, wherein Q satisfies:

Q=Max(QFval)×

QtoA+1.

12. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving a search result obtained in response to a query, wherein the search result identifies a resource and has an associated score S;

computing a recent impression probability for the query for a recent time period and an overall impression probability for the query for an overall time period,

wherein each impression probability corresponds to a ratio of (i) a count of search result impressions selected by users to (ii) a count of all search result impressions presented to users in the respective time periods,

wherein the search result impressions were impressions provided in response to the query by a search engine during the respective time period, and

wherein the overall time period is a time period that began before the recent time period and is longer than the recent time period;

computing a QtoA ratio of the recent impression probability to the overall impression probability;

determining that users prefer newer resources over older resources for the query based on the QtoA ratio;

determining that the resource is a new resource; and

associating a new score S′

with the resource in place of S, wherein the new score S′

signifies a better result than the score S signifies, based on determining, that users prefer newer resources over older resources for the query and that the resource is a new resource.

13. The system of claim 12, wherein the overall time period includes the recent time period.

14. The system of claim 12, wherein the recent time period includes at least one previous day, month, or year.

15. The system of claim 12, wherein a search result impression is a search result that is selected by a user or ranked above a search result that is selected by a user.

16. The system of claim 12, wherein S′

has a boosted value that is higher relative to boosted scores of other resources that are older than the resource.

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

determining a query value Q for the query, the query value being based on the QtoA ratio;

determining a resource value D for the resource, the resource value being based on a function of an age of the resource, F(age), wherein age is the age of the resource; and

calculating S′

as a product of S and a value of a boosting function of the query value Q and the resource value D.

18. The system of claim 17, wherein the boosting function is a continuous function having a range of values that includes values less than 1, a value equal to 1, and values greater than 1, whereby the product of S and the value of the boosting function can be less than S, equal to S, or greater than S.

19. The system of claim 17, wherein the boosting function has the form f (Q, D)=Q^D.

20. The system of claim 18, wherein determining Q comprises determining Q from a maximum Max(QFval) of one or more of the following query freshness values:

a percentile value of a number of occurrences of the query being received within a first recent time period;

percentile value of a number of occurrences of the query on blog web pages within a second recent time period;

a percentile value of a number of occurrences of the query on news web pages within a third recent time period;

a percentile value of a number of occurrences of the query on social network web pages within a fourth recent time period;

a percentile value of a number of occurrences of the query requesting news search results within a fifth recent time period;

a percentile value of a ratio of a number of occurrences of the query requesting news search results within a sixth recent time period to a number of occurrences of the query requesting web search results within the sixth recent time period;

a percentile value of a number of user selections of news search results provided in response to the query within a seventh recent time period; and

a percentile value of a ratio of a number of user selections of news search results provided in response to the query within an eighth recent time period to a number of user selections of web search results provided in response to the query within the eighth recent time period.

21. The system of claim 20, wherein Q satisfies:

22. The system of claim 20, wherein Q satisfies:

Q=Max(QFval)×

QtoA+1.

23. A computer-implemented method comprising:

receiving a first search result obtained in response to a query, wherein the first search result identifies a resource and has an associated score S;

computing a first probability that a particular search result provided in response to the query in a recent time period will be selected when a particular user is presented with an opportunity to select the particular search result;

computing a second probability that the particular search result provided in response to the query in an overall time period will be selected when the particular user is presented with an opportunity to select the particular search result;

computing a QtoA ratio of the first probability to the second probability;

determining, based on the QtoA ratio, that users in the recent time period preferred newer resources for the query than users in the overall time period;

determining that the resource is a new resource; and

applying a boosting function to S to compute a new score S′

for the resource, wherein S′

has a boosted value that is higher relative to boosted scores of other resources that are older than the resource.

24. The method of claim 23, wherein the first probability is a recent impression probability for the query for the recent time period, wherein the recent impression probability corresponds to a ratio of (i) a count of search result impressions selected by users in the recent time period to (ii) a count of all search result impressions presented to users in the recent time period.

25. The method of claim 24, wherein the second probability is an overall impression probability for the query for the overall time period, wherein the overall impression probability corresponds to a ratio of (i) a count of search result impressions selected by users in the overall time period to (ii) a count of all search result impressions presented to users in the overall time period.

26. The method of claim 25, wherein a search result impression is a search result that is selected by a user or ranked above a search result that is selected by a user in a ranked list of search results provided for the query.

27. The method of claim 23, wherein determining that users in the recent time period preferred newer resources for the query than users in the overall time period based on the QtoA ratio comprises determining that the QtoA ratio satisfies a threshold.

28. The method of claim 23, wherein determining that the resource is a new resource comprises determining that an age of the resource satisfies a threshold.

29. The method of claim 23, further comprising:

determining a query value Q for the query, wherein the query value is based on the QtoA ratio;

determining a resource value D for the resource, the resource value being based on a function of an age of the resource, F(age); and

wherein the boosting function is a boosting function of the query value Q and the resource value D.

30. The method of claim 29, wherein the boosting function has the form f (Q, D)=Q^D.

31. The method of claim 29, wherein the function of an age of the resource, F(age), is a monotonically decreasing function of an age of the resource.

32. The method of claim 29, wherein determining Q comprises determining Q from a maximum Max(QFval) of one or more of the following query freshness values:

a percentile value of a number of occurrences of the query being received within a first recent time period;

percentile value of a number of occurrences of the query on blog web pages within a second recent time period;

a percentile value of a number of occurrences of the query on news web pages within a third recent time period;

a percentile value of a number of occurrences of the query on social network web pages within a fourth recent time period;

a percentile value of a number of occurrences of the query requesting news search results within a fifth recent time period;

a percentile value of a ratio of a number of occurrences of the query requesting news search results within a sixth recent time period to a number of occurrences of the query requesting web search results within the sixth recent time period;

a percentile value of a number of user selections of news search results provided in response to the query within a seventh recent time period; and

a percentile value of a ratio of a number of user selections of news search results provided in response to the query within an eighth recent time period to a number of user selections of web search results provided in response to the query within the eighth recent time period.

33. The method of claim 32, wherein Q satisfies:

34. The method of claim 32, wherein Q satisfies:

Q=Max(QFval)×

QtoA+1.

35. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving a first search result obtained in response to a query, wherein the first search result identifies a resource and has an associated score S;

computing a first probability that a particular search result provided in response to the query in a recent time period will be selected when a particular user is presented with an opportunity to select the particular search result;

computing a second probability that the particular search result provided in response to the query in an overall time period will be selected when the particular user is presented with an opportunity to select the particular search result;

computing a QtoA ratio of the first probability to the second probability;

determining, based on the QtoA ratio, that users in the recent time period preferred newer resources for the query than users in the overall time period;

determining that the resource is a new resource; and

applying a boosting function to S to compute a new score S′

for the resource, wherein S′

has a boosted value that is higher relative to boosted scores of other resources that are older than the resource.

36. The system of claim 35, wherein the first probability is a recent impression probability for the query for the recent time period, wherein the recent impression probability corresponds to a ratio of (i) a count of search result impressions selected by users in the recent time period to (ii) a count of all search result impressions presented to users in the recent time period.

37. The system of claim 36, wherein the second probability is an overall impression probability for the query for the overall time period, wherein the overall impression probability corresponds to a ratio of (i) a count of search result impressions selected by users in the overall time period to (ii) a count of all search result impressions presented to users in the overall time period.

38. The system of claim 37, wherein a search result impression is a search result that is selected by a user or ranked above a search result that is selected by a user in a ranked list of search results provided for the query.

39. The system of claim 35, wherein determining that users in the recent time period preferred newer resources for the query than users in the overall time period based on the QtoA ratio comprises determining that the QtoA ratio satisfies a threshold.

40. The system of claim 35, wherein determining that the resource is a new resource comprises determining that an age of the resource satisfies a threshold.

41. The system of claim 35, wherein the operations further comprise:

determining a query value Q for the query, wherein the query value is based on the QtoA ratio;

determining a resource value D for the resource, the resource value being based on a function of an age of the resource, F(age); and

wherein the boosting function is a boosting function of the query value Q and the resource value D.

42. The system of claim 41, wherein the boosting function has the form f (Q, D)=Q^D.

43. The system of claim 41, wherein the function of an age of the resource, F(age), is a monotonically decreasing function of an age of the resource.

44. The system of claim 41, wherein determining Q comprises determining Q from a maximum Max(QFval) of one or more of the following query freshness values:

a percentile value of a number of occurrences of the query being received within a first recent time period;

percentile value of a number of occurrences of the query on blog web pages within a second recent time period;

a percentile value of a number of occurrences of the query on news web pages within a third recent time period;

a percentile value of a number of occurrences of the query on social network web pages within a fourth recent time period;

a percentile value of a number of occurrences of the query requesting news search results within a fifth recent time period;

a percentile value of a ratio of a number of occurrences of the query requesting news search results within a sixth recent time period to a number of occurrences of the query requesting web search results within the sixth recent time period;

a percentile value of a number of user selections of news search results provided in response to the query within a seventh recent time period; and

a percentile value of a ratio of a number of user selections of news search results provided in response to the query within an eighth recent time period to a number of user selections of web search results provided in response to the query within the eighth recent time period.

45. The system of claim 44, wherein Q satisfies:

46. The system of claim 44, wherein Q satisfies:

Q=Max(QFval)×

QtoA+1.