Large-Scale Multimodality Attribute Reduction With Multi-Kernel Fuzzy Rough Sets

A Scalable Approach to Fuzzy Rough Nearest Neighbour Classification with Ordered Weighted Averaging Operators

“…Zeng et al [22,23] present a mechanism to incrementally update fuzzy rough approximations in a hybrid information system (HIS) (in which a hybrid metric combines different types of attributes) and apply this to feature selection. Finally, Hu et al [8] present a distributed implementation of multi-kernel attribute reduction using kernelised fuzzy rough sets, and evaluate the results for Support Vector Machines (SVM) and Classification and Regression Trees (CART).…”

“…As can be seen in Table 1, half of these works only use datasets with up to a few thousand instances. The connected studies of [1,18,2] work with generated datasets of up to 10,000,000 instances and only [8] tests on real datasets with more than one million instances. The studies mentioned above have demonstrated the usefulness of scalable implementations of fuzzy rough prototype and feature selection.…”

“…Since the application of fuzzy rough sets in machine learning problems is a relatively recent, ongoing endeavour, it is not surprising that while there exist distributed implementations of kNN [13] and Fuzzy kNN [12] classification, no Big Data implementation exists of a fuzzy rough set classifier. The few implementations that do try to extend the use of fuzzy rough sets to a Big Data context focus on preprocessing algorithms like Fuzzy Rough Feature and Prototype Selection, and only one has been applied to a real dataset with more than 1 million instances [8].…”

A Fast Feature Selection Algorithm by Accelerating Computation of Fuzzy Rough Set-Based Information Entropy

“…Here, C = max{|U x i B | : x i ∈ U * B }. Therefore, according to Properties 1 and 2, we can use Equation (12) to compute H λ (D|B ∪ {a}) and then obtain an accelerated algorithm in the following.…”

“…Rough set theory [1] presented by Pawlak in 1982 is a useful tool to deal with vagueness and uncertainty information in the field of computer sciences. The research of rough set theory has mainly focused on both the generalizations of rough set models and the applications in different data environments, which has already attached much attention in granular computing [2][3][4], feature selection [5][6][7][8], dynamic data mining [9][10][11], and big data mining [12,13]. On the other hand, since the information entropy is powerful to measure information uncertainty, it has been extensively applied in practical problems, such as decision making [14], time series [15], portfolio selection [16], and so on.…”