LRFM- based association rule mining for dentistry services patterns identification (Case study: a dental center in Iran)

Abstract

Abstract Dentistry processes include prevention, examination of symptoms, and treatment of oral diseases. Since there are various dental services, exploring the combination of services can help both dentists and patients for planning accurately to follow the treatment process in an appropriate order. The aim of this study is to extract the different dental services’ frequent rules. An integrated LRFM, K-means, and APRIORI approach is proposed and implemented in Python programming language. Furthermore, patients’ characteristics and the services provided to patients for an Iranian dental center as a case study are collected. Customers are first divided into 5 categories via LRFM analysis considering the number of referrals, duration of referrals, duration of the last visit, and the total service fee. Subsequently, they are clustered based on features including age, type of insurance, referrer, and group of services received in six categories. Subsequently, in each cluster, there are patients from several groups (according to the LRFM analysis in the previous step). Finally, the sequential rules for dental services are extracted in each cluster and several scenarios are proposed to dental center managers. Results indicate that the rules of dental services can lead to finding some treatment procedures for special cluster of patients in order to remind them of their subsequent referral. The proposed approach provides a better patient treatment process and results in more profits for service providers.