Yazar "Xu, Suping" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    Dual-Channel Fuzzy Interaction Information Fused Feature Selection With Fuzzy Sparse and Shared Granularities
    (IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 14.11.2024) Ju, Hengrong; Fan, Xiaoxue; Ding, Weiping; Huang, Jiashuang; Xu, Suping; Yang, Xibei; Pedrycz, Witold
    Fuzzy information granularity is an effective granular computation approach for feature evaluation and selection. However, most existing methods rely on a single granulation channel, neglecting different granularity representations. In this article, a novel dual-channel fuzzy interaction information fused feature selection with fuzzy sparse and shared granularities is proposed. It mainly comprises the following three parts. First, a dual-channel framework is introduced to construct the fuzzy information granularity from two different strategies. One channel employs sparse mutual strategy to form the sparse representation-based fuzzy information granularity, while the other constructs the fuzzy shared information granularity with a novel fuzzy semi-ball. Second, in each channel, the criteria of maximum relevancy, minimum redundancy, and maximum interaction is adopted to access feature correlation and perform feature ranking. Third, the two feature sequences derived from the dual-channel are fused to form a final feature sequence based on the within-class and between-class mechanism. To validate the efficacy of the proposed method, experimental validations on 15 datasets and schizophrenia data are conducted. The results show that the proposed method outperforms other algorithms in classification accuracy and statistical analysis. Moreover, its superiority regarding accuracy can be demonstrated in the experiments of schizophrenia detection, where it performs well in recognizing schizophrenia through visual interpretation.
  • Küçük Resim
    Öğe
    Incomplete label distribution learning via label correlation decomposition
    (Elsevier b.v., 2025) Xu, Suping; Shang, Lin; Shen, Furao; Yang, Xibei; Pedrycz, Witold
    Label distribution learning (LDL) has garnered increased attention in recent studies on label ambiguity. However, collecting complete annotations for LDL tasks is often time-consuming and labor-intensive compared to traditional learning paradigms. Therefore, designing effective incomplete LDL algorithms is crucial to broaden LDL's application scope. In this paper, we propose a novel LDL algorithm, called Incom plete L abel D istribution L earning via L abel C orrelation D ecomposition (IncomLDL-LCD), which simultaneously learns label distributions and recovers missing description degrees of labels through label correlations. Specifically, we decompose the label correlation into sparse local label correlation and low-rank global label correlation using a softthresholding operator and a singular value thresholding operator, respectively. The former is utilized to capture the related label subsets necessary for reconstructing each possible label, while the latter focuses on extracting the coarse-grained semantic concepts from all labels and exploring the groupings of labels. Additionally, we develop an alternating solution with the accelerated proximal gradient descent method for optimization. Extensive experiments on 16 real-world data sets with varying degrees of missing annotations validate that our algorithm effectively handles incomplete LDL tasks and outperforms state-of-the-art algorithms.