Correlation Based Feature Selection Using Quantum Bio Inspired Estimation of Distribution Algorithm

摘要: Correlation based feature Selection (CFS) evaluates different subsets on the pairwise features correlations and features-class correlations. Machine learning techniques are applied to CFS help in discovering most possible differnt combinations of especillay large spaces. This paper introduces a quantum bio inspired estimation distribution algorithm (EDA) for CFS. The proposed integrates computing concepts, vaccination process with immune clonal selection (QVICA) EDA. It is employed as search technique find optimal subset from space. implemented evaluated using benchmark dataset KDD-cup99 compared GA algorithm. obtained results showed ability QVICA-with EDA obtain better fewer length, higher fitness values reduced computation time.