Abstract | ||
---|---|---|
Cancer is the second most threatening disease in the world today, not only because of its mortality rate, but also due to the brutal changes it imposes on the patient's life, and the fact that its exact causes of progression remain to be discovered. Recent evolution in computer technology has resulted in the emergence of a combined approach to the diagnosis and prognosis process, with a data driven analytical approach complementing biomedical and clinical methods. Cost-sensitive learning is one such data mining method, particularly well suited for medical problems. This paper investigates the performance of a new system based on a hybrid cost-sensitive algorithm (ProICET) on a prostate cancer medical dataset, while trying to produce new medical knowledge. The target of such a system is to reduce the total cost while keeping a high classification accuracy. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1177/1460458208096558 | Health Informatics Journal |
Keywords | Field | DocType |
cost-sensitive learning,hybrid algorithm,data mining,implementation,prostate cancer | Data science,Disease,Hybrid algorithm,Data-driven,Knowledge management,Medical knowledge,Risk analysis (engineering),Prostate cancer,Medicine,Total cost,Cancer,Computer technology | Journal |
Volume | Issue | ISSN |
14 | 4 | 1460-4582 |
Citations | PageRank | References |
0 | 0.34 | 6 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Camelia Vidrighin Bratu | 1 | 0 | 0.68 |
Rodica Potolea | 2 | 34 | 23.33 |
Ioana Giurgiu | 3 | 213 | 14.09 |
Mihai Cuibus | 4 | 4 | 1.12 |