Practical use of data mining techniques in the health insurance system
DOI:
https://doi.org/10.31489/2023ec1/152-161Keywords:
Data Mining, clustering, forecasting, linear regression, polynomial regression, algorithm quality metric, machine mind, insuranceAbstract
Information has become the most important factor in the success of the company's business. The level of competitiveness of the company depends on its availability, quality and timeliness. At the same time, today the amount of incoming information is huge and the need arises for its effective processing. Object: to apply mathematical ways of processing information using the Data Mining toolkit to predict future health insurance benefits.
Methods: to predict the size of future insurance payments and to identify the factors of the occurrence of insurance claims, we chose the methods of linear and polynomial regression of the algorithm metric. The key factors used were: gender and age of customers, smoking habit, weight-to-height ratio, place of residence, presence of children with existing insurance, size of medical expenses and place of residence. Findings: Age and smoking are most important for the level of occurrence of an insured health event. This is what can have the greatest negative impact on the well-being of customers of the insurance company. Body mass index is also quite important. Least of all, the quality of health is influenced by the place of residence, and age. All this allows the management of the insurance company to more clearly plan their own expenses for the payment of insurance compensation. Also, the identification of the degree of influence on the likelihood of an insured event allows the insurance company to more flexibly draw up contracts, which will describe the mechanisms for compensation for the presented damage to the client.
Conclusions: Data Mining can become an effective tool for effective management of the company's activities. Based on the review and comparative analysis of tools and existing approaches to the organization of the intelligent analytical data processing process, a system of criteria and classification of analytical tools was tested to further predict the amount of health damage insurance benefits.
