Optimization model for Sub-feature selection in data mining
Hemanta Kumar Bhuyan, Loveleen Kumar, K. Ramakrishna Reddy · 2019 International Conference on Smart Systems and Inventive Technology (ICSSIT) · 2019
This paper proposes sub-feature selection based classification in data mining. Many existing feature selection methods are used to analyze traditional feature values as well as class values. But all feature's values don't support to distinguish the unique class from existing datasets, because these selected features are irrelevant towards traditional class. Thus it needs to analyze the feature's tiny data such as sub-feature data which help to find distinguish class. So this paper has considered identifying the new class from usual class. It selects needed sub-features from existing feature's data using optimization model. In this paper, it proposed Lagrangian functional method of optimization model for sub-feature selection. This method finds appropriate sub-feature data based on optimal evaluation value. It demonstrates the experiment using different dataset to get sub-feature data for novel class.