Hybrid Multilabel Feature Selection Using BPSO and Neighborhood Rough Sets for Multilabel Neighborhood Decision Systems

Abstract

Recently, feature selection for multilabel classification has attracted substantial attention in many fields; however, some of the available methods ignore the correlations among labels and yield low classification performance. In addition, most feature selection algorithms that are based on multilabel neighborhood rough sets (MNRS) can only deal with finite sets for multilabel data. To address the issues, this paper presents a novel hybrid filter-wrapper multilabel feature selection method that is based on binary particle swarm optimization (BPSO) and MNRS with the Lebesgue measure for multilabel neighborhood decision systems. First, to overcome the problem that the traditional correlation-based feature selection (CFS) algorithm ignores the dependencies among labels, two types of average correlation between single labels and label sets and among labels are presented. Via combination with information-entropy-based uncertainty measures, a new average correlation among labels is studied. A novel comprehensive evaluation function of CFS (NCFS) is constructed. Then, NCFS is introduced as a fitness function into the original BPSO and improved BPSO algorithms to optimize multilabel classification in the early and later stages, respectively, and the optimization process is terminated when the maximum number of iterations is reached. Next, the Lebesgue measure of the neighborhood class is developed for investigating the neighborhood approximation accuracy and the dependency degree based on MNRS. Various properties are deduced, and the relationships among these measures are used to evaluate the uncertainty and correlations among labels of multilabel data. Finally, a hybrid filter-wrapper feature selection algorithm using NCFS-BPSO is designed for preliminarily eliminating redundant features to decrease the complexity, and a heuristic forward multilabel feature selection algorithm is proposed for improving the performance of multilabel classification. Experimental results on fifteen multilabel datasets demonstrate that our proposed algorithms are effective in selecting significant features and realizing great classification performance in multilabel neighborhood decision systems.

A Multi-Domain Diagnostics Approach for Solenoid Pumps Based on Discriminative Features

“…As reported by the authors, wavelet packets energy spectra were extracted as fault features-normal state, small exhaust valve clearance, big exhaust valve clearance, mild leakage, and serious leakage respectively. Also, in [46], a hybrid filter-wrapper feature selection algorithm using a novel comprehensive evaluation function of correlation-based feature selection (NCFS) and BPSO was proposed for improving the performance of multilabel classification problems. Koumi et al [47] combined five filtration methods with different weights to produce a new hybrid filter wrapper algorithm using BPSO and compared its accuracy, performance, and stability with stand-alone filterbased and wrapper-based feature selection algorithms, respectively.…”

Multilabel Feature Selection Using Relief and Minimum Redundancy Maximum Relevance Based on Neighborhood Rough Sets

“…In this paper, the multilabel neighborhood decision system can be simplified as MNDT = <U , C ∪ D>. For any sample x ∈ U and δ ≥ 0, the neighborhood of x is expressed [23] as…”

“…, l t }, where L ∈ D; D j denotes a sample set with label l j , and D i is a label set associated with x i . The lower and upper approximations of D with respect to B are expressed [23] as…”

“…Yu et al [22] designed the variable precision NRS-based multilabel classification algorithm with local correlation. Sun et al [23] proposed the NRS-based multilabel feature selection algorithm with BPSO in neighborhood decision systems. However, these NRS-based feature selection methods still manually determined the neighborhood radius, which easily leads to randomness and uncertainty.…”

Multilabel Feature Selection Using Mutual Information and ML-ReliefF for Multilabel Classification

“…Multilabel classification is an important task of attaching multiple related labels to multilabel instances [9]. It mainly relies on the problem transformation and algorithm adaptation to be solved [10], [11].…”