 |
English
|
正體中文
|
简体中文
|
Items with full text/Total items : 94286/110023 (86%)
Visitors : 21687613
Online Users : 452
|
|
|
Loading...
|
Please use this identifier to cite or link to this item:
http://asiair.asia.edu.tw/ir/handle/310904400/3414
|
| Title: | A Study on the Early Warning System for Health Insurance of Life Insurance Industry |
| Authors: | Chun-Pu Cheng |
| Contributors: | Department of Business Administration |
| Keywords: | Health Insurance;Back-propagation Network;Early Warning System;Moral Hazard |
| Date: | 2004 |
| Issue Date: | 2009-11-17 11:16:56 (UTC+0) |
| Publisher: | Asia University |
| Abstract: | Recent years involve with moral hazard imposturous insurance claim cases are base on injury and life insurance. Since year 1994?s National Health Insurance Program?s start, begin to lead all compatriots to use medical care service?s demand, every of the insurance companies is using this demand to provide all kinds of health care insurance programs, because of large amount of health care insurance plans in the market, so buying health care insurance is common practice. For moral hazard medical care?s claim payments are not great claim or risk not apparent or insurance company?s employees do not have enough experience, pectorals? different, so quality of emphasize are much lower than injury and life insurance cases. The objective for this research paper is using objective science knowledge early warning system in multitudinous medical insurance claim cases, to select out involve in moral hazard?s cases fast and correct and keep insurance system impartial.
This research tries to use one major insurance company-C?s actual medical care?s claim for research data, it takes insurance basic information, insurance accident information and medical information for caution sign structure, and select sixteen variables, apply into Back-propagation Network (BPN) Learning early warning system, and compare the influence of all variables on moral hazard by sensitivity analysis.
The result of experiment has been found the correct predict rate of the samples on four kinds of network structure are up to 98.78%, and the network II (Delta-Rule, Sigmoid) on valuation of RMSE is the best, RMSE is just 0.0784. ?Accident reason? of the sixteen variables is the most sensitive for moral hazard by using sensitivity analysis, the next is ?face amount variation?, and the third is ?medical accident daily benefit?. |
| Appears in Collections: | [經營管理學系 ] 博碩士論文
|
All items in ASIAIR are protected by copyright, with all rights reserved.
|
