Techniques for personalizing expertise related searches

Patent metadata

Field	Value
Publication number	US-10042939-B2
Application number	US-201414529541-A
Country	US
Kind code	B2
Filing date	Oct 31, 2014
Priority date	Oct 31, 2014
Publication date	Aug 7, 2018
Grant date	Aug 7, 2018

Abstract

Disclosed in some examples are methods, systems, and machine-readable mediums which provide for a personalized expertise searching. When a user of the social networking service enters a search query, the system determines if the user is searching for members who possess a particular skill. If the user is searching for members who possess a particular skill, the search results are post-processed by personalizing the search results using one or more machine-learning models which utilize one or more observed features about the user that enters the query, the skills of the members of the social networking service, and the query itself. In some examples, the system may utilize multiple machine-learning models in multiple passes to fine tune the relevance of the search results and to ensure that the post-processing returns search results in a timely manner.

First claim

What is claimed is:

1. A method comprising:
using one or more computer processors to perform operations of:
receiving a query from a computing device over a computer network submitted by a searching user via a search input in a first graphical user interface;
responsive to receiving the query, automatically:
finding in a database of a social networking service a first set of matching member profiles that match one or more keywords in the query;
determining that a keyword in the query is a skill keyword that corresponds to a skill;
responsive to determining that a keyword in the query is the skill keyword:
determining first relevance scores for respective member profiles of the first set of matching member profiles, the first relevance scores reflecting a rough estimate of relevance of the respective profile to a context of the searching user and calculated using a first machine-learned, supervised learning relevance model and a first set of features;
selecting a second set of member profiles comprising member profiles of the first set of matching profiles that have corresponding relevance scores above a predetermined threshold;
determining second relevance scores for respective member profiles for the second set of member profiles, the second relevance scores reflecting a refined relevance of the respective profile to a context of the searching user and calculated using a second machine-learned supervised relevance model using a second set of features different from the first set of features, the second machine-learned supervised relevance model being more computationally expensive than the first machine-learned supervised relevance model;
providing to the computing device for display on the computing device in a second graphical user interface, information about a plurality of the second set of member profiles that is sorted according to the second relevance scores; and
wherein the first and second set of features comprise one or more of: features of the searching user obtained from a member profile of the searching user, one or more keywords from the query, a comparison result of a comparison between an item on the member profile of the searching user and an item on the respective member profiles, and a skill rating of the skill for the respective member profile.

2. The method of claim 1 , wherein the first relevance scores are calculated for all profiles in the first set of matching member profiles.

3. The method of claim 1 , comprising:
determining that a second keyword in the query is a skill keyword that corresponds to a second skill; and
responsive to determining that a second keyword in the query is a skill keyword utilizing a combined skill rating for the skill and the second skill as input into the first relevance model.

4. The method of claim 1 , wherein determining that a keyword in the query is a skill keyword that corresponds to the skill comprises determining that the keyword matches a term corresponding to the skill by string matching.

5. The method of claim 1 , wherein features of the searching user comprise a social network distance between the searching user and a member of the social networking service corresponding to the respective member profile.

6. The method of claim 1 , wherein the features comprise a number of endorsements for the skill that a member of the social networking service corresponding to the respective member profile has.

7. A social networking system comprising:
a processor;
a memory, comprising instructions, that when executed by the processor, causes the processor to perform operations comprising:
receiving a query from a computing device over a computer network submitted by a searching user via a search input in a first graphical user interface;
responsive to receiving the query, automatically:
finding in a database of a social networking service a first set of matching member profiles that match one or more keywords in the query;
determining that a keyword in the query is a skill keyword that corresponds to a skill;
responsive to determining that a keyword in the query is the skill keyword:
determining first relevance scores for respective member profiles of the first set of matching member profiles, the first relevance scores reflecting a rough estimate of relevance of the respective profile to a context of the searching user and calculated using a first machine-learned, supervised learning relevance model and a first set of features;
selecting a second set of member profiles comprising member profiles of the first set of matching profiles that have corresponding relevance scores above a predetermined threshold;
determining second relevance scores for respective member profiles for the second set of member profiles, the second relevance scores reflecting a refined relevance of the respective profile to a context of the searching user and calculated using a second machine-learned supervised relevance model using a second set of features different from the first set of features, the second machine-learned supervised relevance model being more computationally expensive than the first machine-learned supervised relevance model;
providing to the computing device for display on the computing device in a second graphical user interface, information about a plurality of the second set of member profiles that is sorted according to the second relevance scores; and

wherein the first and second set of features comprise one or more of: features of the searching user obtained from a member profile of the searching user, one or more keywords from the query, a comparison result of a comparison between an item on the member profile of the searching user and an item on the respective member profiles, and a skill rating of the skill for the respective member profile.

8. The system of claim 7 , wherein the operations further comprise:
determining that a second keyword in the query is a skill keyword that corresponds to a second skill; and
responsive to determining that the second keyword in the query is a skill keyword, utilizing a combined skill rating for the skill and the second skill as input into the first relevance model.

9. The system of claim 7 , wherein the first relevance model is a neural network.

10. The system of claim 7 , wherein features of the searching user comprise a social network distance between the searching user and a member of the social networking service corresponding to the respective member profile.

11. The system of claim 7 , wherein the features comprise a number of endorsements for the skill that a member of the social networking service corresponding to the respective member profile has.

12. The system of claim 7 , wherein the features comprise an indicator of a common employer between the searching user and a member of the social networking service corresponding to the respective member profile.

13. The system of claim 7 , wherein the features comprise an indicator of a common educational institution between the searching user and member of the social networking service corresponding to the respective member profile.

14. A non-transitory machine-readable medium that stores instructions which when performed by a machine, cause the machine to perform operations comprising:
using one or more computer processors to perform the operations of:
receiving a query from a computing device over a computer network submitted by a searching user via a search input in a first graphical user interface;
responsive to receiving the query, automatically:
finding in a database of a social networking service a first set of matching member profiles that match one o

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Sinha Shakti Dhirendraji
• Thuc Viet-Ha
• Venkataraman Ganesh
• Rodriguez Mario Sergio

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/9536

  Search customisation based on social or collaborative filtering · CPC title

• G06Q50/01

  Physics · mapped topic

• G06F17/30867 Primary

  Physics · mapped topic