Non-negative multi-label feature selection with dynamic graph constraints

Zhang, Yao; Ma, Yingcang

doi:10.1016/j.knosys.2021.107924

Cited by 39 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both [10,11] used dynamic graphs to learn the basic manifold structure of samples or labels and then combined them with linear regression to build feature selection models. Reference [10] strengthens the local connection between samples and labels by combining with subspace to better special features.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Reference [10] strengthens the local connection between samples and labels by combining with subspace to better special features. Reference [11] strengthened the correlation between the weight matrix and sample space and between the weight matrix and label space by comprehensively restricting the weight matrix, making the weight matrix more representative of the weight of features and more accessible to distinguish features.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Optimization of the Intelligent Sports Health Management System Based on Big Data Technology and Internet of Things

Dang

2022

Journal of Function Spaces

View full text Add to dashboard Cite

People often suffer from unpredictable injuries during physical exercise. One of the important reasons is the absence of a scientific sports health management system. Therefore, the construction of such a scientific and effective system has gradually attracted the attention of scholars, which is of great significance to realizing people’s scientific and personalized physical fitness. An intelligent sports health management system based on big data analysis and the Internet of things (IoT) is constructed to solve this problem. The system consists of the user, IoT, cloud, system analysis, evaluation, and data layers. Firstly, a new multilabel feature selection algorithm is proposed in the system analysis layer. The suggested multilabel feature selection algorithm maps the sample space to the label space through the L 21 norm. Then, the consistency of various topologies is guaranteed by combining with feature popularity so that the factors affecting user health can be better selected. Secondly, the experiment is compared with SCLS, SSFS, and six other multilabel feature selection algorithms in 6 classic medical multilabel datasets. Experimental results under five indexes show the effectiveness and superiority of the proposed feature selection algorithm. Finally, the feasibility of the proposed intelligent sports management system is analyzed.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Optimization of the Intelligent Sports Health Management System Based on Big Data Technology and Internet of Things

Dang

2022

Journal of Function Spaces

View full text Add to dashboard Cite

show abstract

“…Firstly, four evaluation criteria are explained. Then we use ten different datasets (Corel5k, Delicious, Flags, Medical, Scene, Enron, GenBase, Social, Yeast, and Emotions) to test CRMIL and compare CRMIL with eight traditional multi-label feature selection algorithms, which are SCLS [ 26 ], D2F [ 30 ], FIMF [ 31 ], PMU [ 3 ], AMI [ 32 ], NMDG [ 33 ], FSSL [ 34 ], and MFS-MCDM [ 35 ].…”

Section: Resultsmentioning

confidence: 99%

Multi-Label Feature Selection with Conditional Mutual Information

Wang

Zhou

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Feature selection is an important way to optimize the efficiency and accuracy of classifiers. However, traditional feature selection methods cannot work with many kinds of data in the real world, such as multi-label data. To overcome this challenge, multi-label feature selection is developed. Multi-label feature selection plays an irreplaceable role in pattern recognition and data mining. This process can improve the efficiency and accuracy of multi-label classification. However, traditional multi-label feature selection based on mutual information does not fully consider the effect of redundancy among labels. The deficiency may lead to repeated computing of mutual information and leave room to enhance the accuracy of multi-label feature selection. To deal with this challenge, this paper proposed a multi-label feature selection based on conditional mutual information among labels (CRMIL). Firstly, we analyze how to reduce the redundancy among features based on existing papers. Secondly, we propose a new approach to diminish the redundancy among labels. This method takes label sets as conditions to calculate the relevance between features and labels. This approach can weaken the impact of the redundancy among labels on feature selection results. Finally, we analyze this algorithm and balance the effects of relevance and redundancy on the evaluation function. For testing CRMIL, we compare it with the other eight multi-label feature selection algorithms on ten datasets and use four evaluation criteria to examine the results. Experimental results illustrate that CRMIL performs better than other existing algorithms.

show abstract

“…en we use ten different datasets (Corel5k, Delicious, Flags, Medical, Scene, Enron, GenBase, Social, Yeast, and Emotions) to test CRMIL and compare CRMIL with eight traditional multi-label feature selection algorithms, which are SCLS [26], D2F [30], FIMF [31], PMU [3], AMI [32], NMDG [33], FSSL [34], and MFS-MCDM [35].…”

Section: Resultsmentioning

confidence: 99%