|
KMID : 1235020180120030001
|
|
Health Service Management Review
2018 Volume.12 No. 3 p.1 ~ p.11
|
|
|
Demographic data based regional clinic revenue estimation model development
|
|
Kim Pil-Jong
Kim Hong-Gee
|
|
|
Abstract
|
|
|
Objectives: The study was conducted to develop the model that predicts the total sales of clinics in city based on population data.
Methods: The clinic revenue, population and population movements data was consolidated into city area unit. The machine learning methods of elastic net and random forest were applied to estimate the clinic revenue and the clinic revenue difference after the five years which was transformed by log and normalization.
Results: Current clinic revenue was most significant factor for the clinic revenue and the clinic revenue difference after the five years prediction. However, clinic revenue difference, the direction between current clinic revenue and future clinic revenue was reversed. The features such as population movement between city and population movement between province also were the important factors for the revenue prediction. Comparing prediction model, the elastic net based machine learning model had more prediction power than random forest based machine learning model.
Conclusion: Based on the results of the study, clinic revenue, population movement between city and population movement between province could be the information factor for the revenue prediction and revenue difference prediction. When considering opening a clinic, demographic-based regional clinic revenue estimation model could be the factors of reference.
|
|
|
KEYWORD
|
|
|
Clinic revenue, Population data, Machine learning, Elastic net, Random forest
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|