Predicting employee loyalty based on machine learning algorithms, Case study: Pars Online Company Call Center

Loyalty is one of the characteristics and benefits admired in all cultures and different historical periods of human beings. From a sociological perspective, loyalty can be considered a positive norm, and loyalty can generally be expressed as dependence on a person or something to which there is a commitment. Organizational loyalty means the high commitment and motivation of employees to the organization, so that they put all their effort and creativity into achieving the goals of the organization. From a macro-management perspective, organizations in today's competitive market consider loyal and committed employees to be one of the main tools for distinguishing themselves from competitors. The costs of employee exit from the organization directly and indirectly result from the fact that organizations make every effort to retain employees and make them loyal to the group. The policy of this research, by looking at the research conducted in the past and drawing out points worth considering from them, is with a new approach, in such a way that high accuracy and confidence in the results obtained, along with low costs, are its main features. The proposed model has benefited from the psychological characteristics of loyalty and the factors affecting employee loyalty from their perspective and their daily issues and concerns in the workplace. The main features of this model are the focus on the welfare status and the elements affecting the working situation and comfort of employees, both from a material and psychological perspective. The model used has presented an acceptable performance in terms of accuracy, due to the diversity and combination of available methods and the multiple use of different algorithms in order to achieve reliable results, and its results can be used with a high degree of confidence to predict the loyalty status of the organization's employees.

کلیدواژه ها:

Employee loyalty ، machine learning ، forecasting