Robust selection of candidates

Patent metadata

Field	Value
Publication number	US-9971976-B2
Application number	US-201414493504-A
Country	US
Kind code	B2
Filing date	Sep 23, 2014
Priority date	Sep 23, 2014
Publication date	May 15, 2018
Grant date	May 15, 2018

Abstract

Selecting candidates from a set of candidates by receiving a set of input parameters about a single aspect of the candidates and applying multiple scoring methods to those input parameters to compute a candidate score via each scoring method. A subset of zero or more candidates is then selected from the set of candidates by applying one or more selection criteria to the set of scores. There is not necessarily a value for every input parameter of every candidate.

First claim

What is claimed is:

1. A method for evaluating candidates for selection, the method comprising:
receiving a first set of input parameters about a first aspect of the candidates in a set of candidates;
applying multiple scoring methods to the first set of input parameters to compute, for each scoring method, a score for the first aspect of each candidate;
evaluating an agreement between or among the multiple scoring methods by comparing the computed scores of the multiple scoring methods;
responsive to the evaluated agreement not meeting or exceeding a threshold level, (i) updating the first set of input parameters by dropping one or more parameters from the first set of input parameters, and (ii) repeating application of the multiple scoring methods with the updated first set of input parameters until the agreement meets the threshold level; and
selecting a subset of one or more candidates from the set of candidates by applying one or more selection criteria to a set of the computed scores corresponding to the agreement meeting the threshold for the first aspect of each candidate;
wherein:
each input parameter in the first set of input parameters is associated with zero or more values for each candidate; and
at least the application of one or more of the multiple scoring methods is performed by computer software running on computer hardware.

2. The method of claim 1 , wherein each scoring method of the multiple scoring methods:
takes as input a same number of input parameters from the first set of input parameters as each other scoring method of the multiple scoring methods; and
takes as input same input parameters from the first set of input parameters as each other scoring method of the multiple scoring methods.

3. The method of claim 1 , further comprising:
computing a score for a second aspect of each candidate using a second set of input parameters;
wherein:
each input parameter in the second set of input parameters is associated with zero or more values for each candidate;
at least one input parameter in either the first set of input parameters or the second set of input parameters is not common to both sets of input parameters; and
the selection of the subset of candidates includes simultaneous evaluation of at least one first aspect score and one second aspect score.

4. The method of claim 3 , wherein the first aspect is candidate capabilities and the second aspect is candidate performance.

5. The method of claim 1 , further comprising:
identifying a selected candidate whose performance could benefit from one or more intervention actions by comparing characteristics of the selected candidate to historic characteristics and performance data about other candidates or other individuals, groups, or entities similarly situated with respect to the compared characteristics; and
recommending the selected candidate for the one or more intervention actions.

6. The method of claim 1 , wherein a first scoring method of the multiple scoring methods includes normalizing each input parameter in the first set of input parameters such that values of the normalized input parameters are comparable.

7. The method of claim 1 , wherein a first scoring method of the multiple scoring methods includes dealing with missing or invalid data in a different way than a second scoring method of the multiple scoring methods.

8. The method of claim 1 , wherein a first scoring method of the multiple scoring methods converts heterogeneous data formats to a homogeneous data format in a different way than a second scoring method of the multiple scoring methods.

9. The method of claim 1 , wherein the candidates are student admissions candidates.

10. A computer program product comprising:
a non-transitory machine readable storage device; and
computer code stored on the non-transitory machine readable storage device, with the computer code including instructions for causing a set of one or more hardware processors to perform operations including the following:
receiving a first set of input parameters about a first aspect of the candidates in a set of candidates;
applying multiple scoring methods to the first set of input parameters to compute, for each scoring method, a score for the first aspect of each candidate;
evaluating an agreement between or among the multiple scoring methods by comparing the computed scores of the multiple scoring methods;
responsive to the evaluated agreement not meeting or exceeding a threshold level, (i) updating the first set of input parameters by dropping one or more parameters from the first set of input parameters, and (ii) repeating application of the multiple scoring methods with the updated first set of input parameters until the agreement meets the threshold level; and
selecting a subset of one or more candidates from the set of candidates by applying one or more selection criteria to a set of the computed scores corresponding to the agreement meeting the threshold for the first aspect of each candidate;

wherein:
each input parameter in the first set of input parameters is associated with zero or more values for each candidate.

11. The product of claim 10 , wherein each scoring method of the multiple scoring methods:
takes as input a same number of input parameters from the first set of input parameters as each other scoring method of the multiple scoring methods; and
takes as input same input parameters from the first set of input parameters as each other scoring method of the multiple scoring methods.

12. The product of claim 10 , further comprising:
computing a score for a second aspect of each candidate using a second set of input parameters;
wherein:
each input parameter in the second set of input parameters is associated with zero or more values for each candidate;
at least one input parameter in either the first set of input parameters or the second set of input parameters is not common to both sets of input parameters; and
the selection of the subset of candidates includes simultaneous evaluation of at least one first aspect score and one second aspect score.

13. The product of claim 10 , further comprising:
identifying a selected candidate whose performance could benefit from one or more intervention actions by comparing characteristics of the selected candidate to historic characteristics and performance data about other candidates or other individuals, groups, or entities similarly situated with respect to the compared characteristics; and
recommending the selected candidate for the one or more intervention actions.

14. The product of claim 10 , wherein a first scoring method of the multiple scoring methods includes normalizing each input parameter in the first set of input parameters such that values of the normalized input parameters are comparable.

15. A computer system for evaluating candidates for selection, the computer system comprising:
a set of one or more hardware processors; and
a computer readable storage medium;
wherein:
computer code, stored on the computer readable storage medium, including instructions for causing the set of one or more hardware processors to perform operations including the following:
receiving a first set of input parameters about a first aspect of the candidates in a set of candidates;
applying multiple scoring methods to the first set of input parameters to compute, for each scoring method, a score for the first aspect of each candidate;
evaluating an agreement between or among the multiple scoring methods by comparing the computed scores of the multiple scoring methods;
responsive to the evaluated agre

Assignees

• IBM

Inventors

• Gifford Wesley M
• Murali Pavankumar
• Sheopuri Anshul

Classifications

• G06Q10/00 Primary

  Administration; Management · CPC title

• G06F16/24578

  using ranking · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G16H50/30

  for calculating health indices; for individual health risk assessment · CPC title

• G06F19/3431

  Physics · mapped topic