Name
Abstract | ||
|---|---|---|
The enormous and increasing cost of health care is burdensome for most low- to middle-income families, especially those families whose members are battling chronic diseases. If effective interventions can be conducted at earlier stages, many costs are avoidable. Correspondingly, predicting the future disease one patient may develop with accuracy is a crucial step towards solving this problem. We have developed a system called CAC, which integrates Clustering, Association analysis and Collaborative filtering to predict patients' future conditions. The data-set used in this study is health insurance data collected from a provincial capital city of China. Specifically, the data-set includes 151,237 insured patients who have reimbursement records between 2007 and 2014. The patients are artificially classified into acute patients and chronic patients. For both sets of patients, we utilise a training set to generate the prediction rules and a testing set to test the prediction results. The results show that for 71% of acute patients and 82% of chronic patients, their future conditions are predictable. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1080/00207543.2016.1208372 | INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH |
Keywords | DocType | Volume |
Health insurance data, data mining, inpatients, personalised disease prediction | Journal | 55 |
Issue | ISSN | Citations |
3 | 0020-7543 | 2 |
PageRank | References | Authors |
0.40 | 13 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Runkang Ding | 1 | 2 | 0.74 |
| Fan Jiang | 2 | 30 | 5.44 |
| Jingui Xie | 3 | 14 | 4.46 |
| Yugang Yu | 4 | 143 | 25.29 |