Dimension Reduction Methodology using Group Feature Selection

Feature selection has become a remarkable research topic in recent years. It is an efficient methodology to tackle the information with high dimension. The underlying structure has been neglected by the previous feature choice technique and it determines the feature singly. Considering this truth, we are going to focus on the matter wherever feature possess some cluster structure. To resolve this downside we are using cluster feature selection technique at cluster level to execute feature choice. Its objective is to execute the feature within the cluster and between the cluster that choose discriminative features and take away redundant options to get optimum subset. We have demonstrate our technique on benchmark knowledge sets and perform the task to attain classification accuracy.