Query-by-example for finding similar people

• US 10,698,914 B2
• Filed: 03/30/2016
• Issued: 06/30/2020
• Est. Priority Date: 03/31/2015
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Qiao, Jian, Nabar, Shubha Umesh, Tunkelang, Daniel
• Primary Examiner(s): Uddin, Mohammed R
• Application Number:
  US15/085,516
• Publication Number:
  US 20160292238A1
• Time in Patent Office: 1,553 Days
• Field of Search: 707767, 707733, 707324, 707759
• CPC Class Codes: G06F 16/242 Query formulation
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/00 User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

Techniques for determining an editable query which, when executed, identifies and presents a set of member profiles similar to a source member profile are described. In an embodiment, a preliminary set of member profiles similar to a source member profile is identified. A set of profile features is obtained from the preliminary set of similar member profiles, and a set of relevant profile features is extracted from the set of profile features. An editable query including the set of relevant profile features is presented. In an embodiment, the set of member profiles similar to the source member profile and corresponding to the set of relevant profile features is also presented. Upon having been edited, the editable query is executable to identify and present a new set of member profiles corresponding to the edited query.

Claims

1. A computer-implemented method comprising:

performing operations for generating and presenting a user interface for editing a compound statement of a plurality of clauses of an editable query to find a first set of member profiles similar to a source member profile, the operations comprising;

receiving a request to determine the first set of member profiles that are similar to the source member profile;

determining the first set of member profiles that are similar to the source member profile, each of the first set of member profiles and the source member profile corresponding to respective members of an electronic social network;

obtaining, from the first set of member profiles, a set of profile features;

extracting, from the set of profile features, a set of relevant profile features, the extracting based on a determination of a usability of each of the set of relevant profile features to generate similarity scores indicating measures of similarity between the source member profile and each of the first set of member profiles;

generating and presenting the user interface, the user interface including a search results portion and a search criteria portion, the search results portion including abbreviated profiles corresponding to the first set of member profiles, the search criteria portion including a plurality of separate user interface controls for separately editing the clauses, wherein the clauses correspond to the set of relevant profile features;

receiving an edit of one of the clauses via one of the plurality of separate user interface controls;

based on a selection of a selectable user interface element, updating the editable query such that it incorporates the edit of the one of the clauses into the compound statement of the plurality of clauses; and

based on a selection of an additional selectable user interface element, updating the search results portion based on an executing the compound statement of the plurality of clauses of the editable search query.

2. The method of claim 1, wherein determining the first set of member profiles comprises:

determining, for a respective member profile in the first set of member profiles, a respective similarity score indicating a measure of similarity between the respective member profile and the source member profile, the respective similarity score determined by;

retrieving a set of profile features from an enhanced member profile corresponding with a respective member profile, the enhanced member profile including profile features extracted from the member profile and enhanced profile features derived based on data in the member profile or retrieved from a data source external to the electronic social network;

deriving a similarity sub-score for each profile feature in the set of profile features, a respective similarity sub-score for a respective profile feature indicating a measure of similarity between the respective profile feature and a corresponding profile feature from the source member profile; and

combining the similarity sub-scores corresponding with each profile feature in the set of profile features to derive the respective similarity score for the respective member profile; and

determining a quantity of member profiles having the highest similarity scores in relation to the source member profile.

3. The method of claim 2, wherein obtaining a set of profile features comprises:

inferring, from at least one member profile within the first set of member profiles, a profile feature for a member profile.

4. The method of claim 2, wherein a profile feature in the set of relevant profile features includes a synonym of the profile feature.

5. The method of claim 4, wherein a synonym of the profile feature is provided in a configuration file.

6. The method of claim 1, further comprising:

presenting a set of target member profiles corresponding to an output of an execution of the editable query.

7. The method of claim 1, wherein the editable query and each of the editable text clauses is editable by at least one of:

text insertion and text deletion.

8. The method of claim 1, wherein a profile feature has a profile feature type of Location, Company, Education, Job Title, Skill, or Seniority.

9. The method of claim 1, wherein extracting includes:

identifying over-indexed profile features from the set of profile features; and

adding the identified over-indexed profiled features into the set of relevant profile features.

10. At least one non-transitory machine-readable medium including instructions which, when executed by a machine, cause the machine to perform operations for generating and presenting a user interface for editing a compound statement of a plurality of clauses of an editable query to find a first set of member profiles similar to a source member profile, the operations comprising:

receiving a request to determine the first set of member profiles that are similar to the source ember profile;

determining the first set of member profiles that are similar to the source member profile;

each of the first set of member profiles and the source member profile corresponding to respective members of an electronic social network;

obtaining, from the first set of member profiles, a set of profile features;

extracting, from the set of profile features;

a set of relevant profile features, the extracting based on a determination of a usability of each of the set of relevant profile features to generate similarity scores indicating measures of similarity between the source member profile and each of the first set of member profiles;

generating and presenting the user interface, the user interface including a search results portion and a search criteria portion, the search results portion including abbreviated profiles corresponding to the first set of member profiles, the search criteria portion including a plurality of separate user interface controls for separately editing each of the clauses, wherein the clauses correspond to the relevant profile features;

receiving an edit of one of the clauses via one of the plurality of separate user interface controls;

based on a selection of a selectable user interface element, updating the editable query such that it incorporates the edit of the one of the clauses into the compound statement of the plurality of clauses; and

based on a selection of an additional selectable user interface element, updating the search results portion based on an executing the compound statement of the plurality of clauses of the editable search query.

11. The at least one non-transitory machine-readable medium of claim 10, wherein the determine the first set of member profiles includes:

determining, for a respective member profile in the first set of member profiles, a respective similarity score indicating a measure of similarity between the respective member profile and the source member profile, wherein to determine the respective similarity score includes to;

retrieving a set of profile features from an enhanced member profile corresponding with a respective member profile, the enhanced member profile including profile features extracted from the member profile and enhanced profile features derived based on data in the member profile or retrieved from a data source external to the electronic social network;

deriving a similarity sub-score for each profile feature in the set of profile features, a respective similarity sub-score for a respective profile feature indicating a measure of similarity between the respective profile feature and a corresponding profile feature from the source member profile; and

combining the similarity sub-scores corresponding with each profile feature in the set of profile features to derive the respective similarity score for the respective member profile; and

determining a quantity of member profiles with the highest similarity scores in relation to the source member profile.

12. The at least one non-transitory machine-readable medium of claim 11, wherein the obtaining of the set of profile features includes inferring, from at least one member profile within the first set of member profiles, a profile feature for a member profile.

13. The at least one non-transitory machine-readable medium of claim 11, wherein a profile feature in the set of relevant profile features includes a synonym of the profile feature.

14. The at least one non-transitory machine-readable medium of claim 13, wherein a synonym of the profile feature is provided in a configuration file.

15. The at least one non-transitory machine-readable medium of claim 10, the operations further comprising:

presenting a set of target member profiles corresponding to an output of an execution of the editable query.

16. The at least one non-transitory machine-readable medium of claim 10, wherein the editable query is editable by at least one of:

text insertion and text deletion.

17. The at least one non-transitory machine-readable medium of claim 10, wherein a profile feature has a profile feature type of Location, Company, Education, Job Title, Skill, or Seniority.

18. The at least one non-transitory machine-readable medium of claim 10, wherein the extracting includes identifying over-indexed profile features from the set of profile features and adding the identified over-indexed profiled features into the set of relevant profile features.

19. An electronic social network comprising:

a non-transitory machine-readable medium including machine-readable instructions which, when executed by one or more processors, cause an electronic social network to perform operations for generating and presenting a user interface for editing a compound statement of a plurality of clauses of an editable query to find a first set of member profiles similar to a source member profile, the operations comprising;

receiving a request to determine the first set of member profiles that are similar to the source member profile;

determining the first set of member profiles that are similar to the source member profile, each of the first set of member profiles and the source member profile corresponding to respective members of an electronic social network;

obtaining, from the first set of member profiles, a set of profile features;

extracting, from the set of profile features, a set of relevant profile features, the extracting based on a determination of a usability of each of the set of relevant profile features to generate similarity scores indicating measures of similarity between the source member profile and each of the first set of member profiles;

generating and presenting the user interface, the user interface including a search results portion and a search criteria portion, the search results portion including abbreviated profiles corresponding to the first set of member profiles, the search criteria portion including a plurality of separate user interface controls for separately editing each of the clauses, wherein the clauses correspond to the relevant profile features;

receiving an edit of one of the clauses via one of the plurality of separate user interface controls;

based on a selection of a selectable user interface element, updating the editable query such that it incorporates the edit of the one of the clauses into the compound statement of the plurality of clauses; and

based on a selection of an additional selectable user interface element, updating the search results portion based on an executing the compound statement of the plurality of clauses of the editable search query.

20. The electronic social network of claim 19, wherein the determining of the first set of member profiles includes:

determining, for a respective member profile in the first set of member profiles, a respective similarity score indicating a measure of similarity between the respective member profile and the source member profile, wherein to determine the respective similarity score includes to;

retrieving a set of profile features from an enhanced member profile corresponding with a respective member profile, the enhanced member profile including profile features extracted from the member profile and enhanced profile features derived based on data in the member profile or retrieved from a data source external to the electronic social network;

deriving a similarity sub-score for each profile feature in the set of profile features, a respective similarity sub-score for a respective profile feature indicating a measure of similarity between the respective profile feature and a corresponding profile feature from the source member profile; and

combining the similarity sub-scores corresponding with each profile feature in the set of profile features to derive the respective similarity score for the respective member profile; and

determining a quantity of member profiles with the highest similarity scores in relation to the source member profile.

21. The electronic social network of claim 20, wherein the obtaining of the set of profile features includes inferring, from at least one member profile within the first set of member profiles, a profile feature for a member profile.

22. The electronic social network of claim 20, wherein a profile feature in the set of relevant profile features includes a synonym of the profile feature.

23. The electronic social network of claim 22, wherein a synonym of the profile feature is provided in a configuration file.