Evaluation of employee profiles using a hybrid clustering and optimization model - Practical study

摘要

Purpose - The purpose of this paper is to introduce an evaluation methodology for employee profiles that will provide feedback to the training decision makers. Employee profiles play a crucial role in the evaluation process to improve the training process performance. This paper focuses on the clustering of the employees based on their profiles into specific categories that represent the employees' characteristics. The employees are classified into following categories: necessary training, required training, and no training. The work may answer the question of how to spend the budget of training for the employees. This investigation presents the use of fuzzy optimization and clustering hybrid model (data mining approaches) as a fuzzy imperialistic competitive algorithm (FICA) and k-means to find the employees' categories and predict their training requirements. Design/methodology/approach - Prior research that served as an impetus for this paper is discussed. The approach is to apply evolutionary algorithms and clustering hybrid model to improve the training decision system directions. Findings - This paper focuses on how to find a good model for the evaluation of employee profiles. The paper introduces the use of artificial intelligence methods (fuzzy optimization (FICA) and clustering techniques (K-means)) in management. The suggestion and the recommendations were constructed based on the clustering results that represent the employee profiles and reflect their requirements during the training courses. Finally, the paper proved the ability of fuzzy optimization technique and clustering hybrid model in predicting the employee's training requirements. Originality/value - This paper evaluates employee profiles based on new directions and expands the implication of clustering view in solving organizational challenges (in TCT for the first time).