Search

In this study, we explore spectral heterogeneity within plant canopies, a characteristic often observed in stressed plants where certain leaves or intra-leaf regions exhibit stress symptoms while others remain unaffected. Considering this variability in spectral signatures holds promise for enhancing remote sensing methodologies aimed at plant stress detection. Typically, remote sensing techniques analyze the plant as a whole, potentially overlooking stress-related spectral signatures due to the inclusion of unaffected pixels. We used a clustering-based technique, which incorporates semi-supervised learning elements for tuning hyper-parameters, to differentiate spectral patterns associated with and unique to pixels from broomrape-infected (Orobanche spp. and Phelipanche spp.) carrots from unrelated patterns. Ground-based hyperspectral (400–1000 nm) images of broomrape-infected and non-infected carrot canopies were used in an agglomerative clustering procedure followed by spectral angle mapper (SAM) analysis to identify a spectral endmember indicative of broomrape infection symptoms. Pixels from this cluster constituted an average of 8.5–11.5 % from the canopies of infected plants. Subsequently, we: (a) examined the relationship between carrot leaf mineral content and the percentage of symptomatic pixels to explore stress-induced alterations creating the unique spectral signatures of infected plants; and (b) utilized the inverse mode of PROSPECT, a radiative transfer model (RTM), to derive primary plant traits from the distinct spectral data of each cluster. We found that deficits in two macro elements, phosphorous and potassium, along with two pigments, chlorophyll and carotenoid, were correlated with the symptomatic cluster in infected plants. The methodology presented in this study paves the way for further research into broomrape detection in various crop species, as well as other plant stressors.

Community detection is a crucial task in network analysis that can be significantly improved by incorporating subject-level information, ie, covariates. Existing methods have shown the effectiveness of using covariates on the low-degree nodes, but rarely discuss the case where communities have significantly different density levels, ie, multiscale networks. In this paper, we introduce a novel method that addresses this challenge by constructing network-adjusted covariates, which leverage the network connections and covariates with a node-specific weight for each node. This weight can be calculated without tuning parameters. We present novel theoretical results on the strong consistency of our method under degree-corrected stochastic blockmodels with covariates, even in the presence of misspecification and multiple sparse communities. Additionally, we establish a general lower bound for the community detection problem when both the network and covariates are present, and it shows that our method is optimal for connection intensity up to a constant factor. Our method outperforms existing approaches in simulations and a LastFM app user network. We then compare our method with others on a statistics publication citation network where 30 % of nodes are isolated, and our method produces reasonable and balanced results. Our method is implemented in the R package NAC. [ABSTRACT FROM AUTHOR]

Core samples analysed through CT scans provide crucial reservoir information for reservoir development, encompassing porosity, permeability, pore size, grain distribution, grain sphericity and roundness, and distribution of their mineralogical assemblages. While 3D CT scans offer detailed insights, their availability is limited. This paper introduces a novel machine learning-based workflow for extracting reservoir properties from 2D CT scan images. The k-means clustering algorithm was applied to segregate and separate the constituent assemblages of the samples and consequently estimating effective porosity, clay volume, and individual pore area of the core samples. Subsequently, we employ a Monte-Carlo simulation on the Kozeny-Carman equation to predict low, mid, and high-case permeabilities of the core samples. The sphericity and roundness of the grains were estimated from the grain assemblage. Furthermore, grain sphericity and roundness were analysed using spectral clustering, another machine learning based algorithm in order to cluster the grains based on these attributes, which help in understanding the depositional transport mechanism of the grains. The feasibility of the proposed method was tested over the sandstone samples from the Lower Fulmar formation (Jurassic) and Leman formation (Permian) of the Anglo-Dutch and Central Graben basins in the North Sea. The results from the study show good correlation and inline with experimental results performed on the core samples. The methodology demonstrated is unique and has novel applications including simulating fluid flow behaviour in porous and permeable media of different constituent assemblages. [ABSTRACT FROM AUTHOR]

Design of an automated and continuous framework is of paramount importance to structural health monitoring (SHM). This study proposes an innovative multi-task unsupervised learning method for early assessment of damage in large-scale bridge structures under long-term monitoring. This method entails three main tasks of data cleaning, data partitioning, and anomaly detection. The first task includes discarding missing data and providing outlier-free samples by developing an approach based on the well-known DBSCAN algorithm. Accordingly, this approach enforces the DBSCAN to generate two clusters, one of which contains outlier-free samples and the other one comprises outlier data. In the second task, the outlier-free samples are fed into spectral clustering to partition them into local clusters. Subsequently, a cluster with the maximum cumulative local density is selected as the optimal partition whose features are extracted as the representative data. Finally, local empirical measures under the theory of empirical learning are used to compute anomaly indices for SHM. Long-term modal frequencies of two full-scale bridges are incorporated to verify the proposed method alongside comparative analyses. Results prove that this method can effectively detect damage by providing discriminative anomaly scores and mitigating the negative influences of severe environmental variability. [ABSTRACT FROM AUTHOR]

Subspace clustering typically clusters data by performing spectral clustering to an affinity matrix constructed in some deterministic ways of self-representation coefficient matrix. Therefore, the quality of the affinity matrix is vital to their performance. However, traditional deterministic ways only provide a feasible affinity matrix but not the most suitable one for showing data structures. Besides, post-processing commonly on the coefficient matrix also affects the affinity matrix's quality. Furthermore, constructing the affinity matrix is separate from optimizing the coefficient matrix and performing spectral clustering, which can not guarantee the optimal overall result. To this end, we propose a new method, affinity adaptive sparse subspace clustering (AASSC), by adding Laplacian rank constraint into a subspace sparse-representation model to adaptively learn a high-quality affinity matrix having accurate p-connected components from a sparse coefficient matrix without post-processing, where p represents categories. In addition, by relaxing the Laplacian rank constraint into a trace minimization, AASSC naturally combines the operations of the coefficient matrix, affinity matrix, and spectral clustering into a unified optimization, guaranteeing the overall optimal result. Extensive experimental results verify the proposed method to be effective and superior. [ABSTRACT FROM AUTHOR]

Recent studies have shown that graph‐based clustering methods are good at processing hyperspectral images (HSIs), while falling short for large‐scale HSIs due to high time complexity. Meanwhile, the performance of these methods relies on the quality of the constructed graph with selected anchor points. More anchor points bring better clustering results for graph‐based methods. Time complexity, however, sees a considerable increase as the number of anchor points grows. Therefore, a method that can obtain efficient clustering accuracy and consumes less time is to be developed. Against this backdrop, a novel algorithm named fast spectral clustering via efficient multilayer anchor graph (FEMAG) is proposed to resolve the accuracy and time‐consuming trade‐off problem. First, FEMAG adopts superpixel principal component analysis (SuperPCA) to extract the low‐dimensional features of HSIs. Then, a multilayer anchor graph is constructed to improve the clustering performance. When constructing the similarity graph, FEMAG takes balanced K‐means‐based hierarchical K‐means (BKHK) to obtain outperforming anchor points efficiently. Extensive experiments validate that FEMAG achieves better clustering accuracy while taking less time compared to previous clustering methods. [ABSTRACT FROM AUTHOR]

Scientific and effective classification of gas wells is conducive to mastering production characteristics of wells and establishing reasonable production measures. Therefore, based on the similarity of time series, and in conjunction with spectral clustering, this paper proposes a new classification model for tight sandstone gas wells: the KF-SDTW-Spectral model. The research results show that the 61 tight sandstone gas wells can be divided into clusters using different classification indicators: two clusters based on monthly gas production, three clusters based on monthly water production, and three clusters based on oil pressure. The model exhibits high clustering quality, surpassing traditional clustering methods. By using this model, the production characteristics of high-yield and low-yield wells are analyzed and corresponding production measures are proposed. This paper provides a new approach on the classification of tight sandstone gas wells, offering valuable guidance for the formulation of production measures in the gas field. [ABSTRACT FROM AUTHOR]

Due to the complex nature and highly heterogeneous of cancer, as well as different pathogenesis and clinical features among different cancer subtypes, it was crucial to identify cancer subtypes in cancer diagnosis, prognosis, and treatment. The rapid developments of high-throughput technologies have dramatically improved the efficiency of collecting data from various types of omics. Also, integrating multi-omics data related to cancer occurrence and progression can lead to a better understanding of cancer pathogenesis, subtype prediction, and personalized treatment options. Therefore, we proposed an efficient multi-omics bipartite graph subspace learning anchor-based clustering (MBSLC) method to identify cancer subtypes. In contrast, the bipartite graph intended to learn cluster-friendly representations. Experiments showed that the proposed MBSLC method can capture the latent spaces of multi-omics data effectively and showed superiority over other state-of-the-art methods for cancer subtype analysis. Moreover, the survival and clinical analyses further demonstrated the effectiveness of MBSLC. The code and datasets of this paper can be found in https://github.com/Julius666/MBSLC. [ABSTRACT FROM AUTHOR]

Multimodal multi-objective problems refer to situations where more than one Pareto set in the decision space corresponds to the same Pareto front in the objective space. In this paper, we propose a multimodal multi-objective coati optimization algorithm based on spectral clustering (MMOCOA-SC) for use in multimodal multi-objective problems. The algorithm starts by creating stable subpopulations via a spectral clustering method, grouping similar individuals within each subpopulation. Then, an improved coati optimization algorithm is proposed in order to increase the diversity of the population and continuously search for near-Pareto optimal solutions. Finally, we use an improved crowding distance method combined with a non-dominated sorting method to maintain and retain multiple near-Pareto optimal solutions. The MMOCOA-SC is evaluated alongside five state-of-the-art algorithms using the 2020 CEC test suite, IDMP test problems, and twelve classic engineering application problems. In comparison to the other five state-of-the-art algorithms, the experimental results show that MMOCOA-SC exhibits a superior performance when applied to the problems. This study explores multimodal multi-objective optimization algorithms with a focus on the concept of symmetry, which is crucial for multimodal multi-objective optimization problems in terms of achieving a balanced decision space and a diversity of solutions in the objective space. [ABSTRACT FROM AUTHOR]

In large-scale simulation optimization, it is impossible to exhaustively simulate every choice. However, there are often inherent similarities between choices: for example, two similar sets of input settings to a simulation model can reasonably be expected to produce similar output. The information gained from simulating one choice can thus be used to infer the values of other similar choices, enabling learning more from a relatively small number of samples. The paper "Sequential Learning with a Similarity Selection Index," by Zhou, Fu, and Ryzhov, develops a new similarity model to improve the final selection decision after all samples have been collected. The new "similarity indices" are complementary to all existing information collection procedures, which do not focus on the final decision. At the same time, the new model allows a tractable theoretical treatment of an optimal procedure, which can be efficiently approximated. We consider the problem of selecting the best alternative in a setting where prior similarity information between the performance output of different alternatives can be learned from data. Incorporating similarity information enables efficient budget allocation for faster identification of the best alternative in sequential selection. Using a new selection criterion, the similarity selection index, we develop two new allocation methods: one based on a mathematical programming characterization of the asymptotically optimal budget allocation and the other based on a myopic expected improvement measure. For the former, we present a novel sequential implementation that provably learns the optimal allocation without tuning. For the latter, we derive its asymptotic sampling ratios. We also propose a practical way to update the prior similarity information as new samples are collected. Numerical results illustrate the effectiveness of both methods. Funding: This work was supported by the Air Force Office of Scientific Research [Grant FA95502010211]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/opre.2023.2478. [ABSTRACT FROM AUTHOR]

Graph Neural Networks (GNNs) have gained popularity in image matching methods, proving useful for various computer vision tasks like Structure from Motion (SfM) and 3D reconstruction. A well-known example is SuperGlue. Lightweight variants, such as LightGlue, have been developed with a focus on stacking fewer GNN layers compared to SuperGlue. This paper proposes the h-GNN, a lightweight image matching model, with improvements in the two processing modules, the GNN and matching modules. After image features are detected and described as keypoint nodes of a base graph, the GNN module, which primarily aims at increasing the h-GNN's depth, creates successive hierarchies of compressed-size graphs from the base graph through a clustering technique termed SC+PCA. SC+PCA combines Principal Component Analysis (PCA) with Spectral Clustering (SC) to enrich nodes with local and global information during graph clustering. A dual non-contrastive clustering loss is used to optimize graph clustering. Additionally, four message-passing mechanisms have been proposed to only update node representations within a graph cluster at the same hierarchical level or to update node representations across graph clusters at different hierarchical levels. The matching module performs iterative pairwise matching on the enriched node representations to obtain a scoring matrix. This matrix comprises scores indicating potential correct matches between the image keypoint nodes. The score matrix is refined with a 'dustbin' to further suppress unmatched features. There is a reprojection loss used to optimize keypoint match positions. The Sinkhorn algorithm generates a final partial assignment from the refined score matrix. Experimental results demonstrate the performance of the proposed h-GNN against competing state-of-the-art (SOTA) GNN-based methods on several image matching tasks under homography, estimation, indoor and outdoor camera pose estimation, and 3D reconstruction on multiple datasets. Experiments also demonstrate improved computational memory and runtime, approximately 38.1% and 26.14% lower than SuperGlue, and an average of about 6.8% and 7.1% lower than LightGlue. Future research will explore the effects of integrating more recent simplicial message-passing mechanisms, which concurrently update both node and edge representations, into our proposed model. [ABSTRACT FROM AUTHOR]

The latest earthquakes (Morrocco, Nepal, Sichuan – China, etc.) have highlighted the critical importance of local-site parameters on the vulnerability of existing building stock. The paper performs the clustering method based on the sub-surface parameters for structural damage prediction. The data set includes the damage status for 44 locations after the 2023 Kahramanmaraş earthquake sequence and local site parameters: Vs30, predominant frequency (f0), horizontal to vertical spectral ratio value (A0), and engineering bedrock depth (VsD760). The Fuzzy C-Means (FCM) and Spectral Clustering (SC) algorithms are carried out on the pre-processed data set, including the sub-surface parameters for each location and the data set clustered into two-clusters within each method. Then, the estimated clusters are compared with the post-earthquake two clusters representing the cluster of damage and no-damage state for considered locations that composed through official damage assessment reports The FCM algorithm yielded a 90% accuracy compared to actual clusters, while the results of the SC algorithm indicated an 86% accuracy. Among the parameters, the VsD760 and f0 demonstrate the ability to establish a discernible demarcation by manifesting distinguishable clustering patterns. Notably, the Area Under the Curve of the Receiver Operating Characteristic (AUC-ROC) value is calculated at 97% and 85% for FCM and SC algorithms, respectively. The outcomes of this study offer the potential to predict the structural damage status of a location under a crucial seismic hazard in the pre-earthquake condition. This enables the development earthquake-resistant cities prior to earthquakes or implement necessary precautions to mitigate seismic risk in the afterward. [ABSTRACT FROM AUTHOR]

Multi-label feature selection can effectively improve the performance and efficiency of subsequent learning tasks by selecting important features within multi-label data. However, for handling multiple labels, many approaches group them to gather insights for label fusion, but ignore the different importance of these label groups and treat them equally, which seems unfair to individual label groups and fails to consider their distinct significances. Moreover, for handling the relationship between features and labels, many multi-label feature selection methods efficiently achieve linear fitting of features and labels through manifold learning, but ignore fitting spatial distribution between feature space and label space. Motivated by these, this paper integrates label distribution learning and spectral clustering to evaluate the unique significance of each label group and construct an improved label space, which is then aligned with the feature space through manifold distribution consistency for multi-label feature selection. First, we propose a hypothetical model indicating the existence of a relationship among labels, wherein this relationship involves clustering subordinate labels around a central core label. On this basis, we employ spectral clustering to generate distinct label clusters by integrating density peaks, thereafter combining this with label distribution learning to assess the significance of each cluster. Then, we design a manifold distribution consistency evaluation, i.e., quantifying the structural disparity between feature space and the enhanced label space achieved through spectral clustering-based label enhancement strategy, so as to obtain a low-dimensional feature space and the optimal feature subset. Finally, experimental results showcase the superiority of our proposed multi-label feature selection algorithm when compared with five other algorithms, across several datasets from diverse domains. [ABSTRACT FROM AUTHOR]

We present a hybrid similarity kernel that exemplifies the integration of short‐ and long‐range descriptors via the use of an average kernel approach. This technique allows for a direct measure of the similarity between amorphous configurations, and when combined with an active learning (AL) spectral clustering approach, it leads to a classification of the amorphous configurations into uncorrelated clusters. Subsequently, a minimum size database is built by considering a small fraction of configurations belonging to each cluster and a machine learning interatomic potential (MLIP), within the Gaussian approximation scheme, is fitted by relying on a Bayesian optimization of the potential hyperparameters. This step is embedded within an AL loop that allows to sequentially increase the size of the learning database whenever the MLIP fails to meet a predefined energy convergence threshold. As such, MLIP are fitted in an almost fully automatized fashion. This approach is tested on two diverse amorphous systems that were previously generated using first‐principles molecular dynamics. Accurate potentials with less than 2 meV/atom root mean square energy error compared to the reference data are obtained. This accuracy is achieved with only 175 configurations sampling the studied systems at various temperatures. The robustness of these potentials is then confirmed by producing models with several thousands of atoms featuring a good agreement with reference ab initio and experimental data. [ABSTRACT FROM AUTHOR]

Attribute reduction is a significant challenge in fields like data mining and pattern recognition. Various models have been introduced to enhance the performance of attribute reduction algorithms, such as the fuzzy rough sets model. However, the common greedy-based reduction algorithm frameworks shared by these models often struggle to efficiently remove redundant attributes. Manual intervention is often employed by researchers to extract the optimal attribute subset, such as setting hyperparameters to control the algorithm’s progression. Unfortunately, these methods lack practical relevance. To address these challenges, this study presents an unsupervised attribute reduction algorithm framework that employs spectral clustering and an attribute significance function. Initially, we introduce an attribute similarity function and a spectral clustering algorithm to capture the data’s main partition structures. We then propose a method for automatically selecting the optimal clustering result, aiming to generate preliminary reduction outcomes. Additionally, we developed a novel unsupervised attribute reduction framework by integrating it with the traditional approach. Furthermore, a specific unsupervised attribute reduction algorithm has been obtained by embedding an unsupervised attribute significance function. Comparative experiments were conducted with six state-of-the-art algorithms across 27 datasets, and the results show that our proposed algorithm demonstrates superior efficiency and effectiveness in attribute selection. [ABSTRACT FROM AUTHOR]

Introduction: The use of machine learning is exploding in all areas of healthcare, including the diagnosis and treatment of heart failure. Supervised machine learning can help predict the onset of heart failure, establish the diagnosis, and even predict decompensations. Conversely, unsupervised machine learning is chiefly used for phenotyping of the heart failure population. Several studies have identified distinctive groups of heart failure patients, but the widespread clinical implementation is still lacking. Aims: Our study aims to identify groups with similar characteristics among patients cared for HFrEF at the City Major Heart and Vascular Clinic of Semmelweis University using unsupervised machine learning and to describe the characteristic features of the resulting groups. We then examine the differences in outcome between the resulting groups. Methods: data from outpatients with reduced left ventricular ejection fraction heart failure were collected in a prospective registry. A total of 27 parameters included anamnestic data, laboratory tests, echocardiographic parameters and EQ5D quality of life questionnaire scores. The composite of hospitalization for heart failure and all-cause mortality was considered as the endpoint of the study. Spectral clustering was used to divide the population into three groups. The groups were plotted spatially using principal component analysis. Finally, we compared the groups in terms of parameters and endpoint occurrence. Results: Three characteristic groups were identified in the analysis of 259 patients. The first group consisted of 89 patients with ischemic etiology, more complaining, renal failure, and requiring duck diuretic therapy. The second group of 99 patients consisted of predominantly younger patients with atrial fibrillation, non-ischemic cardiomyopathy, dilated left ventricle, and a lower ejection fraction, almost exclusively on ARNI therapy. The third group of 71 patients included patients with the best ejection fraction, frequently taking ACE inhibitors and MRAs, and not requiring loop diuretics. Group 1 had significantly worse prognosis than group 2 (p=0.013) with a trend to worse prognosis compared to group 3. Conclusion: Our study identified three groups of patients with different characteristics based on a prospectively managed HFrEF registry. After validation on a larger patient cohort, our data may provide a basis for developing targeted treatment strategies.

Let { A n } n be a sequence of square matrices such that size (A n) = d n → ∞ as n → ∞. We say that { A n } n has an asymptotic spectral distribution described by a Lebesgue measurable function f : D ⊂ R k → C if, for every continuous function F : C → C with bounded support, lim n → ∞ ⁡ 1 d n ∑ i = 1 d n F (λ i (A n)) = 1 μ k (D) ∫ D F (f (x)) d x , where μ k is the Lebesgue measure in R k and λ 1 (A n) , ... , λ d n (A n) are the eigenvalues of A n. In the last decades, the asymptotic spectral distribution of increasing size matrices has become a subject of investigation by several authors. Special attention has been devoted to matrices A n arising from the discretization of differential equations, whose size d n diverges to ∞ along with the mesh-fineness parameter n. However, despite the popularity that the topic has reached nowadays, the role of the so-called test functions F appearing in the above limit relation has been inexplicably neglected so far. In particular, a natural question such as "Which is the largest set of test functions F for which the above limit relation is satisfied?" is still unanswered. In the present paper, we provide a definitive answer to this question by identifying the largest set of test functions F for which the above limit relation is satisfied. We also present some applications of this result to the analysis of spectral clustering, including new asymptotic estimates on the number of outliers. A special attention is devoted to the so-called "inner outliers" of Hermitian block Toeplitz matrices. We conclude the paper with an interpretation of the main result in the context of the vague convergence of probability measures. [ABSTRACT FROM AUTHOR]

Among the nonparametric methods of estimating the number of communities (K) in a community detection problem, methods based on the spectrum of the Bethe Hessian matrices ( H ζ with the scalar parameter ζ) have garnered much popularity for their simplicity, computational efficiency, and robustness to the sparsity of data. For certain heuristic choices of ζ, such methods have been shown to be consistent for networks with N nodes with a common expected degree of ω ( log N) . In this article, we obtain several finite sample results to show that if the input network is generated from either stochastic block models or degree-corrected block models, and if ζ is chosen from a certain interval, then the associated spectral methods based on H ζ is consistent for estimating K for the sub-logarithmic sparse regime, when the expected maximum degree is both o ( log N) and ω (1) , under some mild conditions even in the situation when K increases with N. We also propose a method to estimate the aforementioned interval empirically, which enables us to develop a consistent K estimation procedure in the sparse regime. We evaluate the performance of the resulting estimation procedure theoretically, also empirically through extensive simulation studies and application to a comprehensive collection of real-world network data. for this article are available online. [ABSTRACT FROM AUTHOR]

Community detection is a crucial aspect in network analysis. Real-world networks are often enriched with attributes providing extensive information for nodes beyond mere topology. Integrating these nodal attributes into community detection for attributed networks poses notable challenges and remains an active research field. In this paper, we propose a novel method that incorporates structural information into fused attributes. This is achieved by defining a fusion similarity between nodes, which is a convex combination of topology similarity, pairwise attribute similarity, and attribute similarity with their immediate neighbors. One advantage of the proposed method is its flexibility in identifying communities in disassortative networks, where nodes exhibit more connections between different groups than within their own group. We employ an iterative spectral clustering technique to discover communities and assess the influence of various attributes within these communities. Our experimental results validate the effectiveness of this approach, demonstrating its utility in leveraging node attributes in diverse simulated and real-world network datasets. [ABSTRACT FROM AUTHOR]

The growing complexity of biological data has spurred the development of innovative computational techniques to extract meaningful information and uncover hidden patterns within vast datasets. Biological networks, such as gene regulatory networks and protein-protein interaction networks, hold critical insights into biological features' connections and functions. Integrating and analyzing high-dimensional data, particularly in gene expression studies, stands prominent among the challenges in deciphering these networks. Clustering methods play a crucial role in addressing these challenges, with spectral clustering emerging as a potent unsupervised technique considering intrinsic geometric structures. However, spectral clustering's user-defined cluster number can lead to inconsistent and sometimes orthogonal clustering regimes. We propose the Multi-layer Bundling (MLB) method to address this limitation, combining multiple prominent clustering regimes to offer a comprehensive data view. We call the outcome clusters "bundles". This approach refines clustering outcomes, unravels hierarchical organization, and identifies bridge elements mediating communication between network components. By layering clustering results, MLB provides a global-to-local view of biological feature clusters enabling insights into intricate biological systems. Furthermore, the method enhances bundle network predictions by integrating the bundle co-cluster matrix with the affinity matrix. The versatility of MLB extends beyond biological networks, making it applicable to various domains where understanding complex relationships and patterns is needed. [ABSTRACT FROM AUTHOR]

Aiming at the problems that, based on data statistical characteristics, the classification and recognition method of driving style was easy to ignore the diversity of driving style during driv¬ing, a classification and recognition method of driving style was proposed based on driving events, pectral clustering and random forest. Experiments were designed to collect driving data, and the data were preprocessed to extract turning events and braking events. After standardization and dimension¬lity reduction, the spectral clustering algorithm was used to cluster the driving style of turning vents and braking events respectively. The entropy weight method was used to obtain the driving tyle weights of each driver, and the accuracy of five machine learning algorithms was compared for riving style recognition. Results show that the accuracy of driving style recognition is as 92.73% ased on random forest, which significantly improves the accuracy of driving style recognition. [ABSTRACT FROM AUTHOR]

The K-means algorithm and its variants are well-known clustering techniques. In actuarial applications, these partitioning methods can identify clusters of policies with similar attributes. The resulting partitions provide an actuarial framework for creating maps of dominant risks and unsupervised pricing grids. This research article aims to adapt well-established clustering methods to complex insurance datasets containing both categorical and numerical variables. To achieve this, we propose a novel approach based on Burt distance. We begin by reviewing the K-means algorithm to establish the foundation for our Burt distance-based framework. Next, we extend the scope of application of the mini-batch and fuzzy K-means variants to heterogeneous insurance data. Additionally, we adapt spectral clustering, a technique based on graph theory that accommodates non-convex cluster shapes. To mitigate the computational complexity associated with spectral clustering's O (n 3) runtime, we introduce a data reduction method for large-scale datasets using our Burt distance-based approach. [ABSTRACT FROM AUTHOR]

To explore the individual differences in visual distraction characteristics when drivers perform secondary tasks, a real-car experimental platform was constructed, and visual distraction data were collected from 30 participants. Four indicators, including total duration of visual distraction, number of visual distractions, average duration of a single visual distraction, and maximum duration of a single visual distraction, were selected for characteristic analysis of the collected data. The spectral clustering algorithm was employed to cluster drivers' distraction styles into cautious, normal, and aggressive types. The results indicate that increased complexity of secondary tasks significantly increases total duration, number of visual distractions, and maximum duration of a single visual distraction. Increased vehicle speed significantly decreases total duration of visual distraction, average duration of a single visual distraction, and maximum duration of a single visual distraction, with the impact on total duration of visual distraction and average duration of a single visual distraction becoming more pronounced with increased task complexity. Significant differences are found in visual distraction characteristics among the three distraction styles of drivers, with cautious and aggressive drivers tending towards multiple short-duration and fewer long-duration operation modes, respectively, during secondary task execution. [ABSTRACT FROM AUTHOR]

Imbalanced data poses a significant challenge in machine learning, as conventional classification algorithms often prioritize majority class samples, while accurately classifying minority class samples is more crucial. The synthetic minority oversampling technique (SMOTE) represents one of the most renowned methods for handling imbalanced data. However, both SMOTE and its variants have limitations due to their insufficient consideration of data distribution, leading to the generation of incorrect and unnecessary samples. This paper, therefore, introduces a novel oversampling algorithm called data distribution and spectral clustering-based SMOTE (DDSC-SMOTE). This algorithm addresses the shortcomings of SMOTE by introducing three innovative data distribution-based improvement strategies: adaptive allocation of synthetic sample quantities strategy, seed sample adaptive selection strategy, and synthetic sample improvement strategy. First, we use the k-nearest neighbor sample labels and the local outlier factor algorithm to remove noisy and outlier samples. Next, we leverage spectral clustering to identify clusters within the minority class and propose a dual-weight factor that considers inter-cluster and intra-cluster distances to allocate the number of synthetic samples effectively, addressing interclass and intraclass imbalances. Furthermore, we introduce a relative position weight coefficient to determine the probability of selecting seed samples within the subcluster, ensuring that important minority samples have higher chances of being sampled. Finally, we improve the SMOTE sample synthesis formula for safer generation. Extensive comparisons on real datasets from the UCI repository demonstrate that DDSC-SMOTE outperforms seven state-of-the-art oversampling algorithms significantly in terms of G-mean and F1-score, presenting a data distribution-focused solution for addressing imbalanced data challenges. [ABSTRACT FROM AUTHOR]

In recent decades, multi-view clustering has received a lot of attention. The majority of previous research has assumed that all instances have complete views or at least one view that includes all instances. However, the incomplete multi-view clustering issue arises because real-world data frequently lack instances in each view. We propose a novel anchor-based sparse subspace incomplete multi-view clustering solution to this issue. Through a unified sparse subspace learning framework, the proposed method learns inter-view anchor-to-anchor and intra-view anchor-to-incomplete affinities and fuses them into a consensus sparse anchor graph, which yields a unified clustering result. Our method outperforms other incomplete multi-view clustering methods in three important ways: (1) it uses a small number of hyperparameters to learn a sparse consensus graph from the data; (2) Because of the anchor-based graph construction, it can process large datasets; (3) It is naturally capable of handling both negative entries and multiple views. Last but not least, extensive experiments show that the proposed method is effective, supporting the claim that it consistently outperforms current clustering methods. [ABSTRACT FROM AUTHOR]

Copyright of Coal Science & Technology (0253-2336) is the property of Coal Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Spectral clustering techniques depend on the eigenstructure of a similarity matrix to assign data points to clusters, so that points within the same cluster exhibit high similarity and are compared to those in different clusters. This work aimed to develop a spectral method that could be compared to clustering algorithms that represent the current state of the art. This investigation conceived a novel spectral clustering method, as well as five policies that guide its execution, based on spectral graph theory and embodying hierarchical clustering principles. Computational experiments comparing the proposed method with six state-of-the-art algorithms were undertaken in this study to evaluate the clustering methods under scrutiny. The assessment was performed using two evaluation metrics, specifically the adjusted Rand index, and modularity. The obtained results furnish compelling evidence, indicating that the proposed method is competitive and possesses distinctive properties compared to those elucidated in the existing literature. This suggests that our approach stands as a viable alternative, offering a robust choice within the spectrum of available same-purpose tools. [ABSTRACT FROM AUTHOR]

Batteries are a fundamental storage component due to its various applications in mobility, renewable energies and consumer electronics among others. Regardless of the battery typology, one key variable from a user's perspective is the remaining energy in the battery. It is usually presented as the percentage of remaining energy compared to the total energy that can be stored and is labeled State Of Charge (SOC). This work addresses the development of a hybrid model based on a Lithium Iron Phosphate (LiFePO4) power cell, due to its broad implementation. The proposed model calculates the SOC, by means of voltage and electric current as inputs and the latter as the output. Therefore, four models based on k-Means, Agglomerative Clustering, Gaussian Mixture and Spectral Clustering techniques have been tested in order to obtain an optimal solution. [ABSTRACT FROM AUTHOR]

Cluster analysis, a pivotal step in single-cell sequencing data analysis, presents substantial opportunities to effectively unveil the molecular mechanisms underlying cellular heterogeneity and intercellular phenotypic variations. However, the inherent imperfections arise as different clustering algorithms yield diverse estimates of cluster numbers and cluster assignments. This study introduces Single Cell Consistent Clustering based on Spectral Matrix Decomposition (SCSMD), a comprehensive clustering approach that integrates the strengths of multiple methods to determine the optimal clustering scheme. Testing the performance of SCSMD across different distances and employing the bespoke evaluation metric, the methodological selection undergoes validation to ensure the optimal efficacy of the SCSMD. A consistent clustering test is conducted on 15 authentic scRNA-seq datasets. The application of SCSMD to human embryonic stem cell scRNA-seq data successfully identifies known cell types and delineates their developmental trajectories. Similarly, when applied to glioblastoma cells, SCSMD accurately detects pre-existing cell types and provides finer sub-division within one of the original clusters. The results affirm the robust performance of our SCSMD method in terms of both the number of clusters and cluster assignments. Moreover, we have broadened the application scope of SCSMD to encompass larger datasets, thereby furnishing additional evidence of its superiority. The findings suggest that SCSMD is poised for application to additional scRNA-seq datasets and for further downstream analyses. [ABSTRACT FROM AUTHOR]

It is important to identify the discriminative features for high dimensional clustering. However, due to the lack of cluster labels, the regularization methods developed for supervised feature selection can not be directly applied. To learn the pseudo labels and select the discriminative features simultaneously, we propose a new unsupervised feature selection method, named GlObal and Local information combined Feature Selection (GOLFS), for high dimensional clustering problems. The GOLFS algorithm combines both local geometric structure via manifold learning and global correlation structure of samples via regularized self-representation to select the discriminative features. The combination improves the accuracy of both feature selection and clustering by exploiting more comprehensive information. In addition, an iterative algorithm is proposed to solve the optimization problem and the convergency is proved. Simulations and two real data applications demonstrate the excellent finite-sample performance of GOLFS on both feature selection and clustering. [ABSTRACT FROM AUTHOR]

Copyright of Electric Power Automation Equipment / Dianli Zidonghua Shebei is the property of Electric Power Automation Equipment Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Guangdong University of Technology is the property of Journal of Guangdong University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The algorithms for phylogenetic reconstruction are central to computational molecular evolution. The relentless pace of data acquisition has exposed their poor scalability and the conclusion that the conventional application of these methods is impractical and not justifiable from an energy usage perspective. Furthermore, the drive to improve the statistical performance of phylogenetic methods produces increasingly parameter-rich models of sequence evolution, which worsens the computational performance. Established theoretical and algorithmic results identify supertree methods as critical to divide-and-conquer strategies for improving scalability of phylogenetic reconstruction. Of particular importance is the ability to explicitly accommodate rooted topologies. These can arise from the more biologically plausible non-stationary models of sequence evolution. We make a contribution to addressing this challenge with Spectral Cluster Supertree, a novel supertree method for merging a set of overlapping rooted phylogenetic trees. It offers significant improvements over Min-Cut supertree and previous state-of-the-art methods in terms of both time complexity and overall topological accuracy, particularly for problems of large size. We perform comparisons against Min-Cut supertree and Bad Clade Deletion. Leveraging two tree topology distance metrics, we demonstrate that while Bad Clade Deletion generates more correct clades in its resulting supertree, Spectral Cluster Supertree’s generated tree is generally more topologically close to the true model tree. Over large datasets containing 10,000 taxa and ∼500 source trees, where Bad Clade Deletion usually takes ∼2 h to run, our method generates a supertree in on average 20 s. Spectral Cluster Supertree is released under an open source license and is available on the python package index as sc-supertree.