PREDICTION BASED ON INTEGRATION OF DECISIONAL DNA AND A FEATURE SELECTION ALGORITHM RELIEF-F

Set of experience knowledge structure (SOEKS) and decisional DNA (DDNA), as a knowledge representation, provide features such as learning from experience, dealing with noisy and incomplete data, making precise decisions, and supporting predictions. In this work, we investigate how the combination of DDNA and SOEKS with feature selection learning algorithm RELIEF-F can improve the quality of predictions. The proposed approach is general and extensible in terms of both designing enhanced learning algorithms and application to other domains.