Shi Yang

Shi Yang

Auto-generated profile, based on publicly available information

Shi Yang

Jiangxi Province Distributed Computation Technology Center, Nanchang, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 July 2025 in Journal of King Saud University Computer and Information Sciences
Verified
Journal Article
Open access
PDF file
FXFuhai Xi
Fuhai Xi
MMMengli Mei
Mengli Mei
SYShi Yang
Shi Yang
Hang GuoHang Guo
Hang Guo
MYMin Yu
Min Yu
In complex pattern recognition scenarios, multi-source information fusion faces critical challenges in handling conflicting evidence. This paper proposes an innovative conflict evidence combination rule based on a multi-dimensional weighted evidence optimization method to address this issue. First, systematic indicator classification and coefficient of variation analysis are conducted to rigorously select key conflict evaluation metrics. This approach effectively addresses the issues of incomplete evaluation dimensions and limited indicator selection observed in prior research. Subsequently, the multi-dimensional weighted evidence optimization method is employed to calculate the relative importance of each piece of evidence. On this basis, conflicting evidence is effectively discounted to reduce conflict and uncertainty among the evidence. Finally, the Dempster combination rule is applied to fuse the discounted conflict evidence for decision-making, achieving more accurate and robust judgments. Furthermore, this paper constructs a decision-level multi-source information fusion model. The model integrates the outputs of three basic classifiers—K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Random Forests (RF)—as multi-source information, and applies the proposed conflict evidence combination rule for fusion. This approach fully leverages the advantages of different classifiers, thereby enhancing the accuracy and robustness of pattern recognition. To verify the effectiveness and practical application potential of the proposed method, this paper conducts empirical analysis on multiple sets of experiments. The experimental results demonstrate that, compared with existing methods, the proposed method exhibits significant advantages in accuracy and robustness, proving its potential application value in addressing pattern recognition problems in complex scenarios.

31 August 2026 in Journal of King Saud University Computer and Information Sciences
Verified
Journal Article
Open access
PDF file
MMMengli Mei
Mengli Mei
JLJiahui Luo
Jiahui Luo
SYShi Yang
Shi Yang
TLTao Liu
Tao Liu
MYMin Yu
Min Yu
Conventional evidential deep learning (EDL) represents evidence strength and vacuity but typically converts the learned evidence into a single-label prediction. This may conceal class ambiguity when multiple classes remain plausible and lead to unreliable precise decisions. To address this problem, this paper proposes an evidential set-valued classification framework, termed EDL-SC. First, the decision space is extended from singleton labels to both singleton and class-pair actions, allowing class ambiguity to be explicitly represented. Second, a set-aware supervision term is incorporated into evidential learning to align the learned evidence with the expanded decision space. Third, a maximum expected utility rule is used to select between precise and set-valued predictions, with a risk preference parameter controlling the trade-off between true-class coverage and prediction imprecision. Extensive experiments demonstrate that EDL-SC consistently improves true-class coverage over precise EDL while maintaining selective set-valued outputs. Compared with randomized Adaptive Prediction Sets, it produces more compact prediction sets and achieves a more favorable utility-based balance between coverage and decision specificity. Further analyses show that MEU inference enables the transition from forced precise predictions to set-valued decisions, while set-aware learning helps refine when such decisions should be issued. Overall, EDL-SC provides a framework that aligns evidential learning with MEU-based set-valued decision making, thereby supporting the explicit representation of class ambiguity and controllable decision making under uncertainty.