Cloud-Cluster: An uncertainty clustering algorithm based on cloud model

Liu, Yue; Liu, Zitu; Li, Shuang; Guo, Yike; Li, Qun; Wang, Guoyin

doi:10.1016/j.knosys.2023.110261

Cited by 16 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cloud model is a kind of uncertainty transformation model between qualitative concepts and quantitative values proposed by Academician Li Deyi in combination with probability theory and fuzzy set theory [17] . The cloud model combines randomness and fuzziness, and expresses the integrity of concepts with three digital features: expectation, entropy and over entropy; expresses the association between randomness and fuzziness through digital feature entropy; and realizes the quantitative transformation of qualitative concepts with a specific generator to reflect the uncertainty of qualitative concepts [17][18][19] .…”

Section: Cloud Modelmentioning

confidence: 99%

Study on the Evaluation Model of Earth-Rock Dam Failure Consequences Based on Combined Weighting and Extensible Cloud Model

Cong,

Zhang,

Yin

et al. 2024

Atlantis Highlights in Engineering

View full text Add to dashboard Cite

The reservoir dam disaster is characterized by its suddenness, and the consequences of dam failure are unbearable for people. In order to accurately understand and determine the severity of the consequences of reservoir dam failure, and to ensure the safety and interests of downstream public, this paper comprehensively evaluates the consequences of small and medium-sized earth-rock dams' failure based on the current research status. In this paper, the combination weighting method is used to calculate subjective and objective weights, which not only compensates for the shortcomings of subjective and objective weights but also considers the advantages of both. To ensure that the information of subjective and objective weights is not lost, the minimum information discrimination principle is adopted to determine the comprehensive weight. By combining the advantages of uncertainty inference of the cloud model and the qualitative and quantitative analysis of extensible matter-element theory, a comprehensive evaluation model of extensible cloud dam failure consequences based on combined weighting is constructed. The effectiveness and applicability of the model are verified by using small and medium-sized reservoirs' data to evaluate the severity of reservoir dam failure consequences.

show abstract

Section: Cloud Modelmentioning

confidence: 99%

Study on the Evaluation Model of Earth-Rock Dam Failure Consequences Based on Combined Weighting and Extensible Cloud Model

Cong,

Zhang,

Yin

et al. 2024

Atlantis Highlights in Engineering

View full text Add to dashboard Cite

show abstract

“…However, it is difficult to extend the clustering algorithm to high dimensions in the validation of functional and qualitative clustering. Liu et al [4] proposed a clustering algorithm for Cloud-Cluster that can characterize the object's vagueness and randomness at the same time in order to preserve uncertain information and describe the clusters as concepts. However, this clustering algorithm cannot solve the problem of optimal selection of clustering centers still has limitations.…”

Section: Related Workmentioning

confidence: 99%

A novel high-dimensional clustering algorithm and its application in financial data forecasting

Qian,

Du,

et al. 2024

Third International Conference on Electronic Information Engineering, Big Data, and Computer Technology (EIBDCT 2024)

View full text Add to dashboard Cite

The purpose of this paper is to study the nonlinear characteristics of stock market price rise and fall, and optimize the characteristics of high-dimensional data for the challenges brought by the sparsity and complexity of high-dimensional data to the traditional clustering algorithms. Explore the impact of the multidimensional characteristics and sparsity of stock data on clustering analysis, and propose a high-dimensional clustering algorithm based on Pareto-optimal Steiner points. Traditional clustering algorithms are deficient in dealing with nonlinear data and spatial clustering, while randomly selected clustering centers are easily interfered by market noise, which affects the accuracy of clustering results. In this paper, more accurate clustering results are achieved by selecting Steiner points with Pareto optimality as clustering centers. And the algorithm is applied to the analysis of the stock market, which can effectively help investors choose the appropriate investment portfolio and has important practical application value. Meanwhile, the distance matrix is constructed as the basis of clustering analysis through the preprocessing of stock data and distance metric. The innovation of this paper is that the Steiner points can better represent the center of the clusters thus effectively reducing the computational complexity and storage overhead and improving the efficiency of the algorithm. And the principle of Pareto optimality is applied to the high-dimensional clustering algorithm, which improves the accuracy of stock market analysis.

show abstract

“…Each algorithm has its own advantages and disadvantages depending on the specific problem. For example, the Cloud-Clustering method can quantify the degree of uncertainty during clustering, whereas the GMM model can recognize patterns such as speaker identification through clustering. , However, both the Cloud-Cluster method and GMM are in their infancy stage in the field of material science and can be more computationally expensive than k -Means due to the additional complexity of handling parameters such as expectation, entropy, hyper-entropy for Cloud-Clustering, means, variances, and mixing coefficients for GMM. This increased the computational cost for large data sets like our case.…”

Section: Unknown Solvent Space Explorationmentioning

confidence: 99%

Integrated Supervised and Unsupervised Machine Learning Approach to Map the Electrochemical Windows Over 4500 Solvents for Battery Applications

Manna,

Pathak

2024

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

The compatibility between solvent electrolytes and high-voltage electrode materials represents a significant impediment to the progress of rechargeable metal-ion batteries. Rapidly identifying suitable solvent electrolytes with optimized electrochemical windows (ECWs) within an extensive search space poses a formidable challenge. In this study, we introduce a combined supervised and unsupervised (clustering) machine learning (ML) approach to discern distinct clusters of solvent electrolytes exhibiting varying ECW ranges. Through supervised machine learning, we have accurately predicted optimal solvent electrolytes with desired ECWs from a vast pool of 4882 solvents. Our ML model boasts superior accuracy compared to previously reported data from density functional theory (DFT). Besides, the exploration of the vast solvent space through K-means clustering (unsupervised approach) yields 11 optimal clusters, each encompassing different solvents characterized by diverse ECW ranges and frequencies. The expedited reduction of solvent space achieved through clustering occurs within a very short time frame and with minimal resource expenditure. Consequently, this method is highly capable of streamlining the subsequent experimental investigations for battery applications, avoiding the need for a trial-and-error approach.

show abstract

Cloud-Cluster: An uncertainty clustering algorithm based on cloud model

Cited by 16 publications

References 35 publications

Study on the Evaluation Model of Earth-Rock Dam Failure Consequences Based on Combined Weighting and Extensible Cloud Model

Study on the Evaluation Model of Earth-Rock Dam Failure Consequences Based on Combined Weighting and Extensible Cloud Model

A novel high-dimensional clustering algorithm and its application in financial data forecasting

Integrated Supervised and Unsupervised Machine Learning Approach to Map the Electrochemical Windows Over 4500 Solvents for Battery Applications

Contact Info

Product

Resources

About