Optimizing user selection for performing tasks in social networks

1. A method comprising:

generating a statistical model for modeling a likelihood of social media user response from data collected from social media using a set of available features;

ranking the social media users in an estimation set based on the statistical model;

placing the ranked social media users in a rank ordered list;

selecting a sub-interval on the rank ordered list for maximizing the response rate; and

determining a unit benefit per user response and a unit cost per requested response.

2. The method of claim 1 , further comprising:

using linear projection of the selected sub-interval for user selection of one or more targeted requests for responses.

3. The method of claim 1 , wherein selecting the sub-interval is based on particular constraints.

4. The method of claim 3 , wherein the particular constraints comprise one of a predetermined minimum number of user requests for responses, and a predetermined maximum number of user requests for responses.

5. The method of claim 1 , wherein the collected data comprises answer response data from posted questions.

6. The method of claim 1 , wherein the statistical model is used to rank social media users that are strangers to one another in an estimation set in the order of response probability.

7. The method of claim 6 , wherein actual response information is available for the estimation set.

8. A method comprising:

generating a statistical model for modeling a likelihood of social media user response from data collected from social media using a set of available features;

ranking the social media users in an estimation set based on the statistical model;

placing the ranked social media users in a rank ordered list;

selecting a sub-interval on the rank ordered list for maximizing an expected net benefit for user selection of one or more targeted requests for responses; and

determining a unit benefit per user response and a unit cost per requested response.

9. The method of claim 8 , wherein expected net benefit comprises a difference between a total benefit for improving user response rate and a total cost for improving the user response rate.

10. The method of claim 8 , wherein selecting the sub-interval is based on particular constraints.

11. The method of claim 10 , wherein the particular constraints comprise one of a predetermined minimum number of user requests for responses, and a predetermined maximum number of user requests for responses.

12. The method of claim 8 , wherein the collected data comprises answer response data from posted questions.

13. The method of claim 8 , wherein the statistical model is used to rank social media users that are strangers to one another in an estimation set in the order of response probability.

14. The method of claim 13 , wherein actual response information is available for the estimation set.

15. The method of claim 8 , further comprising:

using linear projection of the selected sub-interval for user selection of one or more targeted requests for responses,

wherein cost per requested response and benefit per user response are determined based on one or more of a linear function and a non-linear function.

16. A computer program product for optimizing user selection for response to requests, the computer program product comprising a computer readable storage medium having program code embodied therewith, the program code readable/executable by a processor to perform a method comprising:

generating a statistical model, by the processor, for modeling a likelihood of social media user response from data collected from social media using a set of available features;

ranking the social media users in an estimation set, by the processor, based on the statistical model;

placing the ranked social media users, by the processor, in a rank ordered list;

selecting a sub-interval on the ranked list based on preselected response related criteria; and

determining, by the processor, a unit benefit per user response and a unit cost per requested response.

17. The computer program product of claim 16 , where the criteria comprises one of maximizing the response rate of one or more targeted requests for user responses, and maximizing an expected net benefit for user selection of one or more targeted requests for responses.

18. The computer program product of claim 17 , further comprising:

using linear projection of the selected sub-interval, by the processor, for user selection of one or more targeted requests for responses, wherein selecting the sub-interval is based on particular constraints comprising one of a predetermined minimum number of user requests for responses, and a predetermined maximum number of user requests for responses.

19. The computer program product of claim 18 ,

wherein the collected data comprises answer response data from posted questions, and the statistical model is used to rank social media users that are strangers to one another in an estimation set in the order of response probability,

wherein actual response information is available for the estimation set, and

wherein the expected net benefit comprises a difference between a total benefit for improving user response rate and a total cost for improving the user response rate.

20. The computer program product of claim 19 ,

wherein cost per requested response and benefit per user response are determined based on one or more of a linear function and a non-linear function.