نوع مقاله : مقاله پژوهشی
نویسندگان
دانشکده اقتصاد، دانشگاه خوارزمی، تهران، ایران.
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
This study analyzing the chain stores’ personnel data intends to predict the employees’ turnover rate and examine its personal and organizational (occupational) determining factors. A dataset of 17542 records including information of personnel's activity status (active or quit) and 12 personal and occupational characteristics in the period of November 2018 to 2014 are used in machine learning algorithms of decision tree and support vector machine (SVM) to predict the workforce turnover rate across the country in a pilot chain store. The results show that job characteristics have a greater effect on employees’ turnover. Specifically, among different characteristics, four factors of personnel transfers, their organizational position (line or staff), years of service and working hours have the most importance and influence on job turnover. Among the individual characteristics, it was also observed that job quit is more prevalent among young people with less than 30 years old. Based on these characteristics, support vector machine (SVM) model with 91% accuracy and F1-score above 90% and decision tree algorithm with 83% accuracy and F1-score have shown a good performance in classification and prediction of employees’ turnover. Based on personal and job characteristics and using data mining and machine learning methods, organizations can set policies to preserve human resources, which will reduce costs and also maintain their competitive advantages necessary for the progress and development of the company.
کلیدواژهها [English]