Feature Subset Selection Based on Variable Precision Neighborhood Rough Sets

Abstract

Rough sets have been widely used in the fields of machine learning and feature selection. However, the classical rough sets have the problems of difficultly dealing with real-value data and weakly fault tolerance. In this paper, by introducing a neighborhood rough set model, the values of decision systems are granulated into some condition and decision neighborhood granules. A concept of neighborhood granular swarm is defined in a decision system. Then the sizes of a neighborhood granule and a neighborhood granular swarm are also given. In order to enhance the fault-tolerant ability of classification systems, we define some concepts of granule inclusion, variable precision neighborhood approximation sets and positive region. We propose a variable precision neighborhood rough set model, and analyze its property. Furthermore, based on the positive region of a variable precision neighborhood, we give the significance of an attribute and use it to select feature subsets. A feature subset selection algorithm to the variable precision neighborhood rough sets is designed. Finally, the feature selection algorithm is carried out on the UCI datasets, and the selected features are tested by the support vector machine (SVM) classification algorithm. Theoretical analysis and experiments show that the proposed method can find the effective and compact feature subsets, which have abilities of fault tolerance.

Modeling and Analysis of New Hybrid Clustering Technique for Vehicular Ad Hoc Network

“…Classical rough sets have the problems of difficulty dealing with real-value data and a low fault-tolerance level. The low tolerance level problem was solved in [9] by proposing a neighborhood rough-sets model to create variable-precision neighborhood approximation sets and positive regions, so the rough set becomes a dynamic model. The dynamic rough set model with the fuzzy scheme was used in [10] to make a more efficient rough set clustering system.…”

Feature selection based on multiview entropy measures in multiperspective rough set

“…The neighborhood radius in the neighborhood rough set is a key factor that directly determines the size of neighborhood granularity for each sample. In existing studies, the neighborhood radius is usually searched at 0.05 intervals in (0, 1] based on the principle that minimizes the number of selected features and maximizes classification accuracy 5–14 . Assuming that the number of selected features is large and the classification accuracy is high when $$\delta =0.5$$, the number of selected features is small and the classification accuracy is not high when $$\delta =0.25$$, at this time, the neighborhood radius value cannot be reasonably selected based on the above principle.…”

An optimized adaptive ensemble model with feature selection for network intrusion detection

“…It inherits the advantages of classical rough set theory, which uses the information of the data itself without any prior knowledge, and it also makes use of the domain space of neighborhood granulation theory, which can directly approximate the continuous numerical features 21 . Related researchers have also conducted a lot of research on neighborhood rough set, 21‐25 which proves its advantages in the field of feature reduction.…”