A5050948151- Complex Network Analysis Techniques
- Multi-Criteria Decision Making
- Opinion Dynamics and Social Influence
- Rough Sets and Fuzzy Logic
- Complex Systems and Time Series Analysis
- Remote-Sensing Image Classification
- Bayesian Modeling and Causal Inference
- Mental Health Research Topics
- Spectroscopy and Chemometric Analyses
- Statistical Mechanics and Entropy
- Data Management and Algorithms
- Advanced Computational Techniques and Applications
- Face and Expression Recognition
- Advanced Decision-Making Techniques
- Computational Drug Discovery Methods
- Fractal and DNA sequence analysis
- Advanced Sensor and Control Systems
- Ecosystem dynamics and resilience
- Machine Learning in Bioinformatics
- Industrial Technology and Control Systems
- Bioinformatics and Genomic Networks
- Advanced Algorithms and Applications
- Polysaccharides and Plant Cell Walls
- Fuzzy Systems and Optimization
- Topological and Geometric Data Analysis
Jiangsu University
2023-2024
South China Agricultural University
2024
Open University of China
2023
The Open University
2023
Ghent University
2017-2022
University of Electronic Science and Technology of China
2018
Southwest University
2014-2016
Nanchang University
2008-2009
The quantification of the complexity network is a fundamental problem in research complex networks. There are many methods that have been proposed to solve this problem. Most existing based on Shannon entropy. In paper, new method which nonextensive statistical mechanics quantify network. On other hand, most single structure factor, such as degree each node or betweenness node. method, both influence and quantified. used constitution discrete probability distribution. centrality entropic...
Supervised hyperspectral image (HSI) classification relies on accurate label information. However, it is not always possible to collect perfectly labels for training samples. This motivates the development of classifiers that are sufficiently robust some reasonable amounts errors in data labels. Despite growing importance this aspect, has been studied literature yet. In paper, we analyze effect erroneous sample probability distributions principal components HSIs, and provide way a...
Decision making is still an open issue in the application of Dempster-Shafer evidence theory. A lot works have been presented for it. In transferable belief model (TBM), pignistic probabilities based on basic probability assignments are used decision making. this paper, multiscale transformation assignment function and plausibility proposed, which a generalization transformation. function, factor q Tsallis entropy to make diversified. An example showing that more reasonable given.
Identifying influential nodes in the complex networks is of theoretical and practical significance. There are many methods proposed to identify networks. In this paper, a local structure entropy which based on degree centrality statistical mechanics identifying network. definition entropy, each node has network, equal The main idea try use influence network replace node's whole identified by intermediate connect those with big value degree. We $Susceptible-Infective$ (SI) model evaluate...
One of the key issues in application Dempster–Shafer evidence theory is transformation between basic probability assignments (BPAs) and probability. In this paper, new based on ordered visibility graph (OVG) proposed to solve decision-making problem. transformation, an OVG can be constructed BPAs. From OVG, network focal elements obtained. The degree a node represents its weight, which essential transformation. Based these weights, results are Some illustrative cases provided demonstrate...
The theory of multi-source information fusion plays a pivotal role in decision-making, especially when handling uncertain or imprecise information. Among the existing frameworks, evidence has proven effective for integrating diverse sources to support informed decision-making. Recently, Random Permutation Set Theory (RPST), an extension theory, shown significant practical value due its ability leverage additional inherent event permutations. This insight inspires utilization permutation data...
BP neural network is easily trapped into the local minimum during training process, which results that it can't get optimal solution, even misjudging in device fault diagnosis. Directing to above problems, a Hopfield-BP diagnosis method was proposed, combined Hopfield network, having global computing ability, with charactering nonlinear classification ability. It avoids be optimum. Implementing new of centrifugal fan has proven pattern recognition could solved well, and accuracy increased...
The structure entropy is an important index to illuminate the property of complex network. Most existing entropies are based on degree distribution But can not illustrate weighted networks. In order study networks, a new networks betweenness proposed in this paper. Comparing with entropy, method more reasonable describe
The structure entropy is one of the most important parameters to describe property complex networks. Most existing struc- ture entropies are based on degree or betweenness centrality. In order networks more reasonably, a new Tsallis nonextensive statistical mechanics proposed in this paper. influence and centrality combined entropy. Compared with entropy, reasonable some situations.
The paper combined the advantage of particle swarm optimization algorithm (PSO), global optimizing ability Hopfield Neural Network, and teacher supervising features BP neural network to construct a new equipment fault diagnosis method with higher precision, compared traditional single network.
Structural analysis in network science is finding the information hidden from topology structure of complex networks. Many methods have already been proposed research on structural networks to find different In this work, sum nodes' betweenness centrality (SBC) used as a new index check how changes process network's growth. We build two four processes growth change will be manifested by SBC. that when are under Barab\'asi-Albert rule, value SBC for each grows like logarithmic function....
The Dempster-Shafer (D-S) evidence theory is extensively utilized for manag- ing uncertain information. Nevertheless, the assessment of dissimilarity between basic probability assignments (BPAs) poses a challenge, influencing clustering, pattern recognition, and improvement fusion rules. This paper presents novel approach grounded in Belief Jensen-Shannon (BJS) divergence, incor- porating penalty coefficient to rectify distance assignments. methodology showcases commendable properties,...