Predicting the Level of Salesperson’s Performance in Encouraging Customers to Use Appropriate Shopping Strategies in Sports Clubs

The customers of a sports club are among the important pillars of its survival. In this paper, with the help of data mining and machine learning methods, a framework is presented to predict the level of effectiveness of salespersons’ performance to encourage customers of clubs to choose an appropriate shopping strategy. This framework uses a set of data on the topics of idealized influence behavior, inspirational motivation behavior, intellectual stimulation behavior, individualized consideration behavior, and smart selling behavior, as its inputs. In the proposed framework, first, the data is refined using the Pearson criterion, and invaluable questions/features are removed from the data set. There are five levels of effectiveness in our questionnaire, and each of them has a different number of records in the data set. So, in the second step, the data set is balanced using repetition, SMOTE, and Int-SMOTE methods. The Int-SMOTE balancing method is introduced in this paper for the first time. It is a SMOTE method with integer outputs. Finally, using different classifiers, we predict the level of effectiveness of salesperson's behaviors in encouraging customers. Evaluating the models indicates that the different models have been able to correctly identify the level of effectiveness of salesperson's behaviors between 76.16% to 96.82%. Also, we confirm our findings about the effects of different salesperson’s behavior to encourage customers using several other published papers.