Deriving and using interaction profiles

• US 10,387,512 B2
• Filed: 12/11/2015
• Issued: 08/20/2019
• Est. Priority Date: 06/28/2004
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: Battle, Alexis J., Cohn, David Ariel, Grimes, Carrie Elizabeth, Lamping, John Ogden
• Primary Examiner(s): Ellis, Matthew J
• Application Number:
  US14/966,545
• Publication Number:
  US 20160098488A1
• Time in Patent Office: 1,348 Days
• Field of Search: None
• 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: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/30 Profiles
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

Systems and methods for deriving and using an interaction profile are described. In one described method, a plurality of metrics indicating a level of satisfaction for search results is determined. The metrics comprise at least one of click-duration data, multiple-click data, and query-refinement data. The values of the metrics for a plurality of instances of an object, such as search results from a search engine, are determined. An interaction profile for the object, based at least in part on the values of the metrics for a plurality of instances of the first object, is then determined. This interaction profile may be used in a variety of ways, such as determining the quality of ranking algorithms and detecting undesirable search results.

Claims

1. A method performed by a system of one or more computers, the method comprising:

identifying a plurality of classes of search queries, each class including a plurality of search queries;

comparing, by the system, for each class of search queries from the plurality of classes, a first variation of a scoring algorithm for scoring electronic documents responsive to search queries in the class with a second variation of the scoring algorithm for scoring the electronic documents responsive to search queries in the class, wherein the comparing includes;

obtaining a first profile that characterizes historical interaction, by a first plurality of users, with first result sets of electronic documents generated in response to search queries in the class, the electronic documents in each first result set scored by a search engine in accordance with the first variation of the scoring algorithm and ranked by the search engine in each first result set in accordance with the scoring,

obtaining a second profile that characterizes historical interaction, by a second plurality of users, with second result sets of electronic documents generated in response to search queries in the class, the electronic documents in each second result set scored by the search engine in accordance with the second variation of the scoring algorithm and ranked by the search engine in each second result set in accordance with the scoring,

determining, from the first profile, one or more first metrics of the historical interaction by the first plurality of users with the first result sets of electronic documents, the one or more first metrics including a first ratio of single clicks to multiple clicks made in the first result sets of electronic documents,

determining, from the second profile, one or more second metrics of the historical interaction by the second plurality of users with the second result sets of electronic documents, the one or more second metrics including a second ratio of single clicks to multiple clicks made in the second result sets of electronic documents,

determining, based on comparison of the one or more first metrics with the one or more second metrics, that the first ratio of single clicks to multiple clicks made in the first result sets of electronic documents is greater than the second ratio of single clicks to multiple clicks made in the second result sets of electronic documents; and

based at least in part on determining that the first ratio of single clicks to multiple clicks made in the first result sets of electronic documents is greater than the second ratio of single clicks to multiple clicks made in the second result sets of electronic documents, optimizing the scoring algorithm to score documents in results sets generated in response to subsequently submitted search queries in the class using the first variation of the scoring algorithm.

2. The method of claim 1, wherein identifying the plurality of classes of search queries comprises:

receiving data identifying a group of queries as one class from the plurality of classes.

3. The method of claim 1, wherein identifying the plurality of classes of search queries comprises:

identifying queries that result in high click duration as one class from the plurality of classes; and

identifying queries that result in medium or low click duration as another class from the plurality of classes.

4. The method of claim 1, wherein:

each class of the plurality of classes includes a plurality of queries;

obtaining the first profile includes;

obtaining, for each query in the plurality of queries, a first query profile that characterizes historical interaction, by the first plurality of users, with electronic documents in first result sets of electronic documents generated in response to the query; and

combining the first query profiles to obtain the first profile; and

obtaining the second profile includes;

obtaining, for each query in the plurality of queries, a second query profile that characterizes historical interaction, by the second plurality of users, with electronic documents in second result sets of electronic documents generated in response to the query; and

combining the first query profiles to obtain the second profile.

5. The method of claim 1, wherein identifying the plurality of classes of search queries comprises:

identifying queries that originated from one country as one class from the plurality of classes; and

identifying queries that originated from another country as another class from the plurality of classes.

6. The method of claim 1, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are navigational queries as one class from the plurality of classes.

7. The method of claim 1, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are commercial queries as one class from the plurality of classes.

8. The method of claim 1, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are category queries as one class from the plurality of classes.

9. The method of claim 1, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are general queries as one class from the plurality of classes.

10. A system comprising:

one or more computers, including one or more processors, programmed to perform operations, the operations comprising;

identifying a plurality of classes of search queries, each class including a plurality of search queries;

comparing, by the system, for each class of search queries from the plurality of classes, a first variation of a scoring algorithm for scoring electronic documents responsive to search queries in the class with a second variation of the scoring algorithm for scoring the electronic documents responsive to search queries in the class, wherein the comparing includes;

obtaining a first profile that characterizes historical interaction, by a first plurality of users, with first result sets of electronic documents generated in response to search queries in the class, the electronic documents in each first result set scored by a search engine in accordance with the first variation of the scoring algorithm and ranked by the search engine in each first result set in accordance with the scoring,

obtaining a second profile that characterizes historical interaction, by a second plurality of users, with second result sets of electronic documents generated in response to search queries in the class, the electronic documents in each second result set scored by the search engine in accordance with the second variation of the scoring algorithm and ranked by the search engine in each second result set in accordance with the scoring,

determining, from the first profile, one or more first metrics of the historical interaction by the first plurality of users with the first result sets of electronic documents, the one or more first metrics including a first ratio of single clicks to multiple clicks made in the first result sets of electronic documents,

determining, from the second profile, one or more second metrics of the historical interaction by the second plurality of users with the second result sets of electronic documents, the one or more second metrics including a second ratio of single clicks to multiple clicks made in the second result sets of electronic documents,

determining, based on comparison of the one or more first metrics with the one or more second metrics, that the first ratio of single clicks to multiple clicks made in the first result sets of electronic documents is greater than the second ratio of single clicks to multiple clicks made in the second result sets of electronic documents; and

based at least in part on determining that the first ratio of single clicks to multiple clicks made in the first result sets of electronic documents is greater than the second ratio of single clicks to multiple clicks made in the second result sets of electronic documents, optimizing the scoring algorithm to score documents in results sets generated in response to subsequently submitted search queries in the class using the first variation of the scoring algorithm.

11. The system of claim 10, wherein identifying the plurality of classes of search queries comprises:

receiving data identifying a group of queries as one class from the plurality of classes.

12. The system of claim 10, wherein identifying the plurality of classes of search queries comprises:

identifying queries that result in high click duration as one class from the plurality of classes; and

identifying queries that result in medium or low click duration as another class from the plurality of classes.

13. The system of claim 10, wherein:

each class of the plurality of classes includes a plurality of queries;

obtaining the first profile includes;

obtaining, for each query in the plurality of queries, a first query profile that characterizes historical interaction, by the first plurality of users, with electronic documents in first result sets of electronic documents generated in response to the query; and

combining the first query profiles to obtain the first profile; and

obtaining the second profile includes;

obtaining, for each query in the plurality of queries, a second query profile that characterizes historical interaction, by the second plurality of users, with electronic documents in second result sets of electronic documents generated in response to the query; and

combining the first query profiles to obtain the second profile.

14. The system of claim 10, wherein identifying the plurality of classes of search queries comprises:

identifying queries that originated from one country as one class from the plurality of classes; and

identifying queries that originated from another country as another class from the plurality of classes.

15. The system of claim 10, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are navigational queries as one class from the plurality of classes.

16. The system of claim 10, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are commercial queries as one class from the plurality of classes.

17. The system of claim 10, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are category queries as one class from the plurality of classes.

18. The system of claim 10, wherein identifying the plurality of classes of search queries comprises:

identifying queries that are general queries as one class from the plurality of classes.

19. A method performed by a system of one or more computers, the method comprising:

identifying a plurality of classes of search queries, each class including a plurality of search queries;

comparing, by the system, for each class of search queries from the plurality of classes, a first variation of a scoring algorithm for scoring electronic documents responsive to search queries in the class with a second variation of the scoring algorithm for scoring the electronic documents responsive to search queries in the class, wherein the comparing includes;

obtaining a first profile that characterizes historical interaction, by a first plurality of users, with first result sets of electronic documents generated in response to search queries in the class, the electronic documents in each first result set scored by a search engine in accordance with the first variation of the scoring algorithm and ranked by the search engine in each first result set in accordance with the scoring,

obtaining a second profile that characterizes historical interaction, by a second plurality of users, with second result sets of electronic documents generated in response to search queries in the class, the electronic documents in each second result set scored by the search engine in accordance with the second variation of the scoring algorithm and ranked by the search engine in each second result set in accordance with the scoring,

generating, from the first profile, one or more first metrics of the historical interaction by the first plurality of users with the first result sets of electronic documents, the one or more first metrics including at least one of a number of clicks made on the first result sets of electronic documents or a number of refinements of search queries that elicited the first result sets of electronic documents,

generating, from the second profile, one or more second metrics of the historical interaction by the second plurality of users with the second result sets of electronic documents, the one or more second metrics including at least one of a number of clicks made on the second result sets of electronic documents or a number of refinements of search queries that elicited the second result sets of electronic documents,

determining, based on comparison of the one or more first metrics with the one or more second metrics, that at least one of;

(i) the number of clicks made on the first result sets of electronic documents is less than the number of clicks made on the second result sets of electronic documents, or

(ii) the number of refinements of search queries that elicited the first results sets of electronic documents is less than the number of refinements of search queries that elicited the second results sets of electronic documents; and

in response to the determination, optimizing the scoring algorithm to score documents in results sets generated in response to subsequently submitted search queries in the class using the first variation of the scoring algorithm.