System and method for finding matches between users in a networked environment

• US 9,251,220 B2
• Filed: 11/20/2013
• Issued: 02/02/2016
• Est. Priority Date: 11/20/2013
• Status: Active Grant

• Current Assignee: Match Group, LLC (Match Group, Inc.)
• Original Assignee: Match.com LLC (Match Group, Inc.)
• Inventors: Klawitter, Monica, Thombre, Amarnath, Lam, Sydney C., Carrico, Todd
• Primary Examiner(s): TRAN, ANHTAI V
• Application Number:
  US14/084,718
• Publication Number:
  US 20150142823A1
• Time in Patent Office: 804 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Finding a match for an end user on an online social platform is not a trivial task. To improve match making, various methods and systems are disclosed which are configured to compute heuristics for various end users, which help to predict the likelihood that two end users would ultimately engage in some form of communication with each other. The heuristics are used in an algorithm (i.e., a statistical/predictive model) for providing a set of matches to an end user. These heuristics may be computed based on varying forms of communication which indicate different levels of engagement between end users, and/or based on some other indication of how an end user may react to another end user.

Claims

1. A method for selecting one or more matching end users for an end user to be matched on an online social platform where end users participate in different forms of electronic communications with varying levels of engagement with other end users, comprising:

determining a first set of end users who have participated in a first form of electronic communication with the end user to be matched;

determining a second set of end users who have participated in the first form of electronic communication with one or more end users in the first set of end users, the second set of end users being end users who are similar to the end user being matched;

determining a third set of end users who have participated in the first form of electronic communication with one or more end users in the second set of end users;

computing a score value for an end user in the third set of end users based on a sum of products, each product is a product of a coefficient and a variable;

wherein the coefficients correspond to effects of the variables on the score value;

wherein the variables used in computing the score value for a particular end user comprises;

a triangulation variable based on the number of end users in the second set of end users with whom the particular end user has participated in the first form of electronic communication; and

one or more search criteria variables based on whether the particular end user has user profile attributes which meets the search criteria of the end user to be matched; and

selecting the one or more matching end users based on the score values computed for the end users in the third set of end users.

2. The method according to claim 1, wherein the coefficients used if the end user to be matched is female differ from the coefficients used for each variable if the end user to be matched is male.

3. The method according to claim 1, wherein the one or more search criteria variables used if an end user to be matched is female, are different from the one or more search criteria variables used if the end user to be matched is male.

4. The method according to claim 1, wherein:

the score value represents a probability that the particular end user will engage in one form of electronic communication with the end user to be matched; and

the sum of products are provided based on a logistic regression model for computing the probability.

5. The method according to claim 1, further comprising:

if any of the first set of end users, the second set of end users, and the third set of end user is an empty set, repeating the determining steps based on participation in a second form of electronic communication, wherein the second form of electronic communication indicates a level of engagement that is lower than the level of engagement indicated by the second form of electronic communication.

6. The method according to claim 5, further comprising:

repeating the computing step using a different sum of products having different coefficients and different variables.

7. The method according to claim 1, wherein the one or more the search criteria variables include:

a distance variable measuring a distance between the location of the particular end user and the location of the end user to be matched;

wherein coefficient corresponding to the distance variable is provided such that the distance variable contributes negatively to the score value.

8. The method according to claim 1, wherein at least one of the one or more the search criteria variables takes into account a positive ratings variable as a trade-off based on whether the particular end user is well received by other end users on the online social platform when determining whether the particular end user has user profile attributes which meets the search criteria of the end user to be matched.

9. The method of claim 1, wherein the third set of end users are selected from a pool of end users who meet one or more of the following requirements:

the end user has not already been rated by the end user being matched;

the end user has not already been selected as a matching end user for the end user being matched in the past; and

the end user has not already participated in a form of electronic communication with the end user being matched in the past.

10. The method of claim 1, wherein at least one of the one or more search criteria variables is relaxed within a buffer when determining whether the particular end user has user profile attributes which meets the search criteria of the end user to be matched.

11. The method of claim 10, wherein the buffer is provided if the particular end user or the end user to be matched meets one or more conditions.

12. The method of claim 10, wherein the buffer is not provided if the particular end user or the end user to be matched meets one or more conditions.

13. One or more non-transitory tangible media that includes code for execution and when executed by a processor is operable to perform operations for selecting one or more matching end users for an end user to be matched on an online social platform where end users participate in different forms of electronic communications with varying levels of engagement with other end users, the operations comprising:

determining a first set of end users who have participated in a first form of electronic communication with the end user to be matched;

determining a second set of end users who have participated in the first form of electronic communication with one or more end users in the first set of end users, the second set of end users being end users who are similar to the end user being matched;

determining a third set of end users who have participated in the first form of electronic communication with one or more end users in the second set of end users;

computing a score value for an end user in the third set of end users based on a sum of products, each product is a product of a coefficient and a variable;

wherein the coefficients correspond to effects of the variables on the score value;

wherein the variables used in computing the score value for a particular end user comprises;

a triangulation variable based on the number of end users in the second set of end users with whom the particular end user has participated in the first form of electronic communication; and

one or more search criteria variables based on whether the particular end user has user profile attributes which meets the search criteria of the end user to be matched; and

selecting the one or more matching end users based on the score values computed for the end users in the third set of end users.

14. The media according to claim 13, wherein the coefficients used if the end user to be matched is female differ from the coefficients used for each variable if the end user to be matched is male.

15. The media according to claim 13, wherein the one or more search criteria variables used if an end user to be matched is female, are different from the one or more search criteria variables used if the end user to be matched is male.

16. The media according to claim 13, wherein:

the score value represents a probability that the particular end user will engage in one form of electronic communication with the end user to be matched; and

the sum of products are provided based on a logistic regression model for computing the probability.

17. A server for selecting one or more matching end users an end user to be matched on an online social platform where end users participate in different forms of electronic communications with varying levels of engagement with other end users, the server comprising:

a memory for storing a log of the different forms of electronic communications occurring through the online social platform; and

a processor for;

determining a first set of end users who have participated in a first form of electronic communication with the end user to be matched;

determining a second set of end users who have participated in the first form of electronic communication with one or more end users in the first set of end users, the second set of end users being end users who are similar to the end user being matched;

determining a third set of end users who have participated in the first form of electronic communication with one or more end users in the second set of end users;

computing a score value for an end user in the third set of end users based on a sum of products, each product is a product of a coefficient and a variable;

wherein the coefficients correspond to effects of the variables on the score value;

wherein the variables used in computing the score value for a particular end user comprises;

a triangulation variable based on the number of end users in the second set of end users with whom the particular end user has participated in the first form of electronic communication; and

one or more search criteria variables based on whether the particular end user has user profile attributes which meets the search criteria of the end user to be matched; and

selecting the one or more matching end users based on the score values computed for the end users in the third set of end users.

18. The system according to claim 17, wherein the coefficients used if the end user to be matched is female differ from the coefficients used for each variable if the end user to be matched is male.

19. The system according to claim 17, wherein the one or more search criteria variables used if an end user to be matched is female, are different from the one or more search criteria variables used if the end user to be matched is male.

20. The system according to claim 17, wherein:

the score value represents a probability that the particular end user will engage in one form of electronic communication with the end user to be matched; and

the sum of products are provided based on a logistic regression model for computing the probability.