Predicting the retention of customers of sport gym using the K-nearest neighbor algorithm

Document Type : Original Article

Author

Associate Professor, Department of Sports Management, Faculty of Physical Education and Sports Sciences, Allameh Tabatabai University, Tehran, Iran.

10.22034/sms.2024.140657.1302

Abstract

The technique of data mining by augmenting the comprehensive data successfully used in different areas enables us to find the hidden knowledge to influence sports services. The focus of this research is on predicting the retention of club customers using the K-nearest neighbor algorithm (KNN). The statistical population of this developmental-applied study was related to 724 athletes who participated in the online invitation (WhatsApp, Instagram, Telegram, etc.) by completing the questionnaire in the present study. After removing the unqualified questionnaires, finally, 537 athletes in the age group of 20 to 60 years participated in the present study. The electronic, anonymous and self-made questionnaire had 75 factors related to customer satisfaction, based on the received feedback, several changes were made in the questionnaire and finally 18 factors were selected as the main factors of drop or retention related to the facilities and conditions of sports clubs. The face validity of the questionnaire was checked by 5 university professors and experts in related fields. The results showed that the KNN algorithm can predict with 73.4% accuracy and 71.6% accuracy the retention percentage of private club customers who repeat purchases. This study showed that by discovering hidden patterns and relationships in the data, it is possible to use this algorithm correctly to improve the quality of management of sports facilities in order to prevent falling and maintain more.

Keywords

Main Subjects

Files

Share

How to cite

Statistics