Multi-label Feature Selection Based on Fuzzy Neighborhood Mutual Discrimination Index

Wang, Chenxi; E, Chen; Ren, Mengli; Yu, Xiehua; Lin, Yaojin; Li, Shaozi

doi:10.1109/itme56794.2022.00130

Cited by 1 publication

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with FNMDI, 25 MRFI is an extension of FNMDI. Relative entropy in MRFI is suitable for measuring the similarity between two probability distributions, and it has been proved that relative entropy deepens the measurement and feature evaluation of uncertainty in feature selection.…”

Section: Compare With Fnmdimentioning

confidence: 99%

“…However, the above approach considers labels to be equally important to instances, without taking into account that the importance of the label description of the sample varies in real life. Therefore, Wang et al 25 proposed a multi-label feature selection algorithm based on fuzzy neighborhood mutual discrimination index (FNMDI), which used label enhancement to transform multi-label data into labeled distribution data with real-value distribution, and used neighborhood joint entropy to measure the correlation of samples under label space.…”

mentioning

confidence: 99%

“…Finally, the fuzzy neighborhood mutual discrimination index is used to combine the candidate features with the label neighborhood similarity matrix to select discriminative features. Compared with FNMDI, 25 the relative entropy in MRFI deepens the uncertain measurement and feature evaluation of feature selection, which is more robust and advanced. As a new label distribution-based feature selection method, the main contributions of the proposed method are as follows:…”

mentioning

confidence: 99%

See 2 more Smart Citations

Multi‐label feature selection based on relative entropy and fuzzy neighborhood mutual discrimination index

Wang

Ren

et al. 2023

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

Multi‐label feature selection eliminates irrelevant and redundant features, and then improves the performance of multi‐label classification models. Most multi‐label feature selection algorithms assume that the training set contains logical labels, which means that labels are equally important for instances. However, in practical applications, there are different importances with respect to labels. To solve the problem, a multi‐label feature selection method based on relative entropy and fuzzy neighborhood mutual discriminant index is proposed. Firstly, logical labels are converted to label distribution through label enhancement. Secondly, the neighborhood and relative entropy are introduced into the label distribution, the label neighborhood similarity matrix is constructed to describe the similarity of samples under label space. Finally, the fuzzy neighborhood mutual discrimination index is used to combine the candidate features with the label neighborhood similarity matrix, which is used to judge the distinguishing ability of the candidate features. Comprehensive experiment of eight multi‐label datasets shows that the proposed algorithm has better classification performance than other compared algorithms.

show abstract

Section: Compare With Fnmdimentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation