Feature preparing for competency-based on team recruitment

Effective team is a factor contributed to project achievement and organizational goals. Generally, the method of selecting and qualifying team member is often based on familiarity and personal relationships, in which the competency difference of each team candidate was not highly prioritized. Thus, this would affect project performance and potential of success. This research aims to propose method in recruiting team members based on the matching of appropriateness between individual competency and organizational expectation and expectable outcomes from each project. The applied feature selection methods were 1) Information Gain (IG) and 2) Chi Squared (CHI) in the selection of the appropriate personal characteristics. To characterize personal basic competencies all three methods such as Naïve Bayes: NB, Support Vector Machine: SVM and Decision Tree: DTREE were adopted. The results show that Feature selection with Information Gain, yielded better consistence but less time reduction compared to the Chi Squared. Moreover, with competency-based qualifying method, there was 8.02% team classification efficiency increased, compared to personal relationship based method. Conclusively, team selection method based on the appropriate feature, with individual competency-based and classification by support vector machine method demonstrates the best performance in which F₁ measured is 88.16%.