Large Scale Multi-Modality Attribute Reduction with Multi-Kernel Fuzzy Rough Sets

Abstract-In the first place, objects are typically characterized by means of multi-modality attributes, including categorical, numerical, text, image, audio and even videos. In these cases, data are usually high-dimensional, structurally complex, and granular. To point out, those attributes exhibit some redundancy and irrelevant information. Under those circumstances, and combination of multi-modality attributes pose great challenges to traditional classification algorithms. And also, multi-kernel learning handles multi-modality attributes by using different kernels to extract information coming from different attributes.since, consider the aspects fuzziness in fuzzy classification Fuzzy rough sets emerge as a powerful vehicle to handle fuzzy and uncertain attribute reduction. Particularly, we design a framework of multi-modality attribute reduction based on multi-kernel fuzzy rough sets. First, a combination of kernels based on set theory is defined to extract fuzzy similarity for fuzzy classification with multi-modality attributes. Then, a model of multi-kernel fuzzy rough sets is constructed. Finally, an efficient attribute reduction algorithm for large scale multi-modality fuzzy classification based on the proposed model.As a result, experimental demonstrate the effectiveness of the proposed model and the corresponding algorithm.