Pages 456–471

Authors

Tanawat Piriyapattana and Chumpol Bunkhumpormpat

Abstract

With the growing need for personalized marketing in aesthetic clinics, understanding customer behavior is essential. This study aims to analyze the relationship between Decision Tree and Association Rule Mining to uncover patterns in service selection. Association Rule Mining identifies frequently co-occurring services, such as customers undergoing "Acne Clear 6 times" often opting for "VPL red marks 3 times" (Confidence = 58%) and those selecting "General Aesthetic Treatments" commonly using "Skin Supplement" (Lift = 17.46). Meanwhile, Decision Tree analysis segments customers based on service preferences, visit days, and demographics, revealing that Acne Treatments and Skin & Meso Rejuvenation are the most common, with Botox & Injectables peaking on Saturdays. The integration of both models confirms that Acne Treatment customers frequently choose Skin Repair or Laser Treatment, aligning with the Decision Tree’s segmentation. This study demonstrates that combining Decision Tree and Association Rule Mining enhances service recommendations, allowing clinics to implement targeted promotions, such as Laser Treatment discounts for Acne Treatment customers. The findings highlight the value of Machine Learning techniques in refining customer segmentation, improving recommendations, and optimizing marketing strategies in aesthetic clinics.