Application Of Fp-Growth And Fp-Tree Algorithms To The Treatment Selection Decision System In Beauty Clinics (Chrisa Aesthetic Clinic Case Study)

Abstract

The beauty industry faces challenges in understanding consumer preferences for skincare services. This study develops a decision support system based on data mining using the FP-Growth and FP-Tree algorithms to analyze treatment selection patterns at Christa Aesthetic Clinic. Customer transaction data were analyzed to identify service associations based on skin types. Results show that customers with acne-prone skin tend to choose acne exfoliation, acne skinbooster, and acne facial treatments. Dull skin is commonly treated with brightening peels, dermapen derma glow, and brightening facials. Sensitive skin customers prefer facial detox, DNA salmon skinbooster, and moisturizers, while normal skin types tend to choose light exfoliation, sunscreen, moisturizers, and facials. The FP-Growth algorithm effectively identified frequent treatment combinations with a recommendation accuracy of 87%. A satisfaction survey revealed that 95% of customers were satisfied with recommendations tailored to their skin needs. This system enhances operational efficiency and customer experience while providing a data-driven foundation for clinics to formulate more targeted and personalized service strategies.