A case-based system to be applied in the Brazilian public health system

Case-based systems constitute an Artificial Intelligence technique, which is well-known for their ability to reuse and to adapt past experiences to solve new problems. To achieve this, they store and retrieve past experiences, based on their attributes, adapt them, and apply the adapted solution to the new problem. After analysing the solution performance, it may store the new case in the case-based, so that the system can improve its performance over time. The main feature of such systems is the retrieving mechanism, which is responsible for comparing and identifying similar, relevant information in the stored cases, through similarity metrics. This work presents an approach for handling multivalued attributes in well-known similarity metrics in order to retrieve the most relevant cases for a specific class of application. A health area application concerning lateral epicondilitis, which is an elbow tendonitis, was developed., and it illustrates its use. This work contains real data from the Brazilian public health system. where the work was developed and is been prepared to be used.