Next generation similar profiles

Patent metadata

Field	Value
Publication number	US-10733243-B2
Application number	US-201715691623-A
Country	US
Kind code	B2
Filing date	Aug 30, 2017
Priority date	Aug 30, 2017
Publication date	Aug 4, 2020
Grant date	Aug 4, 2020

Abstract

A system, a machine-readable storage medium storing instructions, and a computer-implemented method described herein are directed to a Similar Profiles Engine. The Similar Profiles Engine generates an inverted index query based on one or more portions of profile data of a target member account of a social network service. The Similar Profiles Engine identifies respective profile data, of one or more candidate member accounts in the social network service, that maps to at least one inverted index filter, the at least one inverted index filter matching at least a portion of the inverted index query. The Similar Profiles Engine calculates a similarity score between each respective candidate member account and the target member account, and causes a display of identifiers of one or more candidate member accounts in a user interface of a client device based on respective similarity scores.

First claim

What is claimed is:

1. A computer system, comprising:
one or more hardware processors; and
a non-transitory machine-readable medium for storing instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
generating an inverted index query based on at least a portion of profile data of a target member account of a social network service;
identifying respective profile data, of one or more candidate member accounts in the social network service, that maps to at least one inverted index filter, by:
identifying a first inverted index filter that matches a first pre-selected type of profile data included in the inverted index query;
identifying in the inverted index one or more candidate member accounts that map to the first inverted index filter;
identifying a second inverted index filter that matches a second pre-selected type of profile data included in the inverted index query; and
identifying in the inverted index one or more candidate member accounts that map to the second inverted index filter; and

generating a short list of the one or more candidate member accounts that map to the first inverted index filter, and the one or more candidate member accounts that map to the second inverted index filter; and
for each respective candidate member account, generating a pairing between the respective candidate member account and the target member account, and calculating a respective similarity score for the pairing.

2. The computer system of claim 1 , wherein the generating of the inverted index query based on at least the portion of profile data of the target member account of the social network service includes:
populating the inverted index query with at least one of the following instances of pre- selected types of profile data of the target member account:
a country code identifier, a profile language identifier, one or more industry identifiers, one or more skill tags, or one or more job title keywords.

3. The computer system of claim 2 , wherein the identifying of the respective profile data includes:
identifying a first inverted index filter that matches a first pre-selected type of profile data included in the inverted index query, the first pre-selected type profile data comprising a presence of a specific job title keyword in a profile section; and
identifying in the inverted index one or more candidate member accounts that map to the first inverted index filter, wherein respective profile data of each of the one or more candidate member accounts includes the presence of the specific job title keyword in the profile section.

4. The computer system of claim 1 , wherein the calculating of the similarity score for the pairing of each respective candidate member account and the target member account includes:
calculating each respective similarity score according to a machine learned logistic regression model.

5. The computer system of claim 4 , wherein the machine learned logistic regression model comprises a plurality of pre-defined features with corresponding regression coefficients, the plurality of pre-defined features including at least one of a date of birth, a graduation year, a company identifier, a job title, a profile headline, or a field of study.

6. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more hardware processors of a machine, cause the machine to perform operations comprising:
generating an inverted index query based on at least a portion of profile data of a target member account of a social network service;
identifying respective profile data, of one or more candidate member accounts in the social network service, that maps to at least one inverted index filter, the at least one inverted index filter matching at least a portion of the inverted index query, by:
identifying a first inverted index filter that matches a first pre-selected type of profile data included in the inverted index query;
identifying in the inverted index one or more candidate member accounts that map to the first inverted index filter;
identifying a second inverted index filter that matches a second pre-selected type of profile data included in the inverted index query; and
identifying in the inverted index one or more candidate member accounts that map to the second inverted index filter; and

generating a short list of the one or more candidate member accounts that map to the first inverted index filter, and the one or more candidate member accounts that map to the second inverted index filter; and
for each respective candidate member account, generating a pairing between the respective candidate member account and the target member account, and calculating a respective similarity score for the pairing.

7. The non-transitory machine-readable storage medium of claim 6 , wherein the generating of the inverted index query based on at least the portion of profile data of the target member account of the social network service includes:
populating the inverted index query with at least one of the following instances of pre- selected types of profile data of the target member account:
a country code identifier, a profile language identifier, one or more industry identifiers, one or more skill tags, or one or more job title keywords.

8. The non-transitory machine-readable storage medium of claim 7 , wherein the identifying of the respective profile data includes:
identifying a first inverted index filter that matches a first pre-selected type of profile data included in the inverted index query, the first pre-selected type profile data comprising a presence of a specific job title keyword in a profile section; and
identifying in the inverted index one or more candidate member accounts that map to the first inverted index filter, wherein respective profile data of each of the one or more candidate member accounts includes the presence of the specific job title keyword in the profile section.

9. The non-transitory machine-readable storage medium of claim 6 , wherein the calculating of the similarity score for the pairing of each respective candidate member account and the target member account includes:
calculating each respective similarity score according to a machine learned logistic regression model.

10. The non-transitory machine-readable storage medium of claim 9 , wherein the machine learned logistic regression model comprises a plurality of pre-defined features with corresponding regression coefficients, the plurality of pre-defined features including at least one of a date of birth, a graduation year, a company identifier, a job title, a profile headline, or a field of study.

11. A computer-implemented method, comprising:
generating an inverted index query based on at least a portion of profile data of a target member account of a social network service;
identifying respective profile data, of one or more candidate member accounts in the social network service, that maps to at least one inverted index filter, the at least one inverted index filter matching at least a portion of the inverted index query; and
calculating, using one or more hardware processors, a similarity score between each respective candidate member account and the target member account by:
identifying a first inverted index filter that matches a first pre-selected type of profile data included in the inverted index query;
identifying in the inverted index one or more candidate member accounts that map to the first inverted index filter;
identifying a second inverted index filter that matches a second pre-selected type of profile data included in th

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Zhou Runfang
• Singh Ajit Paul
• Nair Anish Ramdas
• Zhou Sen
• Tcheprasov Vladislav
• Hosmani Sachin
• Teng Da

Classifications

• G06Q10/40

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

• G06Q10/42 Primary

  Determination of affinities or common interests between users · CPC title

• G06F16/951 Primary

  Indexing; Web crawling techniques · CPC title

• G06Q50/01

  Physics · mapped topic