| Summary: | Case-Based Reasoning (CBR) is a reasoning system that uses old knowledge to solve new problem. The new problems solved by adopting the solution of the similar old problems. Knowledge in CBR will increase when the new case is stored in the case base. Scope of CBR applications are very widely, shown with its implementations on variety of different fields. For example at law fields, engineering, computing, communications networks, financial, including the medicine fields. This research is implemented of CBR to diagnose cardiovascular disease is based on the calculation of the similar features in the old case. Features that are used to match the old cases with new cases were age, gender, risk factors and symptoms. Diagnose process is done by entering the case features into the system, then the system will look for cases that have features in common with the new case (retrieve) tures that are used to perform matching old cases with new cases were age, gender, risk factors and symptoms. Any similar cases will be calculated using the weighted Minkowski methode. Cases with a high degree of similarity in most cases will be adopted as a new solution. If similarity value <0.8 then it will be revised by experts. The test results show that the expert who carried out the system is able to diagnose correctly. While the test is based on using medical records data, shows that the system is able to recognize correctly I21 disease (sensitivity) at 100%, instead of I21 recognize disease (specificity) of 83.33%, with an accuracy rate of 95.83% and the errors rate of 4.17%.
|
|---|