Inconsistency guided robust attribute reduction

Attribute reduction (AR) plays an important role in reducing irrelevant and redundant domain attributes, while maintaining the underlying semantics of retained ones. Based on Earth Mover’s Distance (EMD), this paper presents a robust AR algorithm from perspective minimising inconsistency between discernibility reduct entire original attribute set. Due to susceptibility gauger noisy information, strategy for instance denoising is also proposed by detecting abnormal local class distributions with regard global distribution. With such pretreatment process AR, robustness found significantly improved, as testified systematic experimental investigations. The results demonstrate that gained approach generally outperforms those attained application popular, state-of-the-art techniques, terms both size classification using reduced attributes.