CLIAMS:1. A computer implemented method to rank human profiles personalized to a selector, the method comprising:
obtaining querying criteria, by a processor (108), to query a database comprising a set of human profiles, wherein each profile in the set of human profiles is associated with a plurality of attributes;
determining, by the processor (108) a subset of human profiles from the set of human profiles based on the querying criteria;
obtaining, by the processor (108) a selection based ranking for the subset of human profiles, wherein the selection based ranking includes selection of a human profile from a pair of human profiles, and wherein the pair of human profiles are selected from the subset of human profiles;
determining, by the processor (108) a ranking function based on the selection based ranking, wherein the ranking function is indicative of an inclination of the selector towards at least one implicit attribute from among the plurality of attributes; and
applying, by the processor (108) the ranking function to rank a fresh set of human profiles personalized to the selector.
2. The method as claimed in claim 1 further comprising determining the querying criteria, wherein the determining comprises:
providing, by the processor (108), at least one query, to the selector, wherein the at least one query corresponds to information associated with at least one explicit attribute from among the plurality of attributes; and
obtaining, by the processor (108), a response on the at least one query from the selector, to determine the querying criteria.
3. The method as claimed in claim 1, wherein the set of human profiles are obtained, by the processor (108), based on extracting the plurality of attributes to a predefined template, from each applicant's profile.
4. The method as claimed in 1 wherein the subset of human profiles may be determined , by the processor (108), based on combining at least two attributes from among the plurality of attributes based on the querying criteria.
5. The method as claimed in claim 1, wherein the plurality of attributes comprise work experience, educational qualification, project experience, project count, role, technology, previous remuneration, skill set, professional certifications, hobbies, and location.
6. The method as claimed in claim 1, wherein the selection based ranking, by the processor (108), is indicative of at least one implicit attribute preferred by the selector while selecting the human profile from the pair of human profiles.
7. The method as claimed in claim 1, wherein the ranking function is determined , by the processor (108), based on employing active learning techniques, and wherein the active learning techniques capture at least one implicit attribute based on the selection based ranking, and wherein the active learning techniques comprising at least one of a SVM technique, boosting technique, and neural network based approach.
8. The method as claimed in claim 1, wherein each of the at least one implicit attribute in the ranking function is associated with a weight, and wherein the weight is indicative of the relevance of the attribute.
9. The method as claimed in claim 1 further comprising evaluating, by the processor (108), the ranking function based on utilizing a sample set of human profiles.
10. A human profile ranking system (102) to rank of human profiles based on selection criteria personalized to a selector, the human profile ranking system (102) comprising:
a processor (108);
a querying module (120) coupled to the processor (108), to determine a subset of human profiles from a set of human profiles based on querying criteria and a default ranking mechanism, wherein each human profile from among the set of human profiles is associated with a plurality of attributes; and
a function generation module (122) coupled to the processor (108), to:
obtain selection based ranking for a subset of human profiles, wherein the selection based ranking includes selection of a human profile from a pair of human profiles, and wherein the pair of human profiles are selected from the subset of human profiles; and
determine a ranking function based on the selection based ranking, and wherein the ranking function is indicative of an inclination of the selector towards the at least one implicit attribute.
11. The human profile ranking system (102) as claimed in claim 10 further comprises an extraction module (118) coupled to the processor (108), to:
extract a plurality of attributes from each applicant's profile to a predefined template to obtain the set of human profiles; and
combine at least two attributes from among the plurality of attributes, based on a querying criteria.
12. The human profile ranking system (102) as claimed in claim 11, wherein the querying module (120) is to further determine the querying criteria, wherein the determining comprises:
providing at least one query, to the selector, wherein the at least one query corresponds to information associated with at least one explicit attribute from among the plurality of attributes; and
obtaining a response on the at least one query from the selector, to determine the querying criteria.
13. The human profile ranking system (102) as claimed in claim 10, wherein the function generation module (122) further evaluates the ranking function to apply the ranking function to rank a fresh set of human profiles
14. The human profile ranking system (102) as claimed in claim 10, wherein the function generation module (122) determines the ranking function based on active learning techniques, and wherein the active learning techniques capture at least one implicit attribute based on the selection decisions from the selection based ranking by ascertaining a statistical model based on a transformation of attributes to a higher dimensional space, and wherein the statistical model is indicative of degree of inclination towards the at least one implicit attribute from the selection based ranking.
15. The human profile ranking system (102) as claimed in claim 10, wherein the function generation module (122) associates a weight for each of the at least one implicit attribute in the ranking function, and wherein the weight is indicative of the relevance of the attribute.
16. A non-transitory computer readable medium having a set of computer readable instructions that, when executed, cause a computing system to:
obtain querying criteria to query a database comprising a set of human profiles, wherein each profile in the set of human profiles is associated with a plurality of attributes;
determine a subset of human profiles from the set of human profiles based on the querying criteria;
obtain selection based ranking for the subset of human profiles, wherein the selection based ranking includes selection of a human profile from a pair of human profiles, and wherein the pair of human profiles are selected from the subset of human profiles;
determine a ranking function based on the selection based ranking, wherein the ranking function is indicative of an inclination of the selector towards at least one implicit attribute from among the plurality of attributes; and

apply the ranking function to rank a fresh set of human profiles personalized to the selector. ,TagSPECI:As Attached