Spectral Clustering Research Articles

Integrating DIA Single-Cell Proteomics Data with the DiagnoMass Proteomic Hub for Biological Insights.

Single-cell proteomics has emerged as a powerful technology for unraveling the complexities of cellular heterogeneity, enabling insights into individual cell functions and pathologies. One of the primary challenges in single-cell proteomics is data generation, where low mass spectral signals often preclude the triggering of MS2 events. This challenge is addressed by Data Independent Acquisition (DIA), a data acquisition strategy that does not depend on peptide ion isotopic signatures to generate an MS2 event. In this study, we present data generated from the integration of DIA single-cell proteomics with a version of the DiagnoMass Proteomic Hub that was adapted to handle DIA data. DiagnoMass employs a hierarchical clustering methodology that enables the identification of tandem mass spectral clusters that are discriminative of biological conditions, thereby reducing the reliance on search engine biases for identifications. Nevertheless, a search engine (in this work, DIA-NN) can be integrated with DiagnoMass for spectral annotation. We used single-cell proteomic data from iPSC-derived neuroprogenitor cell cultures as a test study of this integrated approach. We were able to differentiate between control and Rett Syndrome patient cells to discern the proteomic variances potentially contributing to the disease's pathology. Our research confirms that the DiagnoMass-DIA synergy significantly enhances the identification of discriminative proteomic signatures, highlighting critical biological variations such as the presence of unique spectra that could be related to Rett Syndrome pathology.

Detection of Evasive Android Malware Using EigenGCN

Recently there is an upsurge in Android malware that use obfuscation and repackaging techniques for evasion. Malware may also combine both these techniques to create stealthy adversarial mimicry samples to launch mimicry attacks. In mimicry attacks, the adversary makes sure that the static and dynamic features present in the crafted malware mimics the features present in the legitimate applications. In such cases, the existing detection mechanisms may become less effective. We found that the malicious nature of Android applications can be determined by identifying certain subgraphs that appear in their system call graphs. These subgraphs can be determined with the help of spectral clustering mechanism present in EigenGCN. With this, the system call graph G will be partitioned into two subgraphs G1 and G2, in which the malicious functionality if any will be present in the subgraph G1. The graph Fourier transform based pooling technique in EigenGCN then computes the features of the subgraphs in the form of graph signals. This graph signals serve as a robust signature to detect malware. The proposed mechanism gave an accuracy of 98.7% on common malware, 97.3% on obfuscated malware, 97.8% on repackaged malware, and 90% on adversarial mimicry malware datasets. As far as we know, this is the first work that proposes a malware detection mechanism, that can detect common as well as obfuscated, repackaged, and mimicry malware in Android.

Microblog Negative Comments Data Analysis Model Based on Multi-scale Convolutional Neural Network and Weighted Naive Bayes Algorithm

As a form of public supervision, Microblog’s negative reviews allow people to share their opinions and experiences and express dissatisfaction with unfair and unreasonable phenomena. This form of supervision has the potential to promote social fairness, drive governments, businesses, and individuals to correct mistakes and enhance transparency. To characterize the sentiment trend and determine the influence of Microblog negative reviews, we propose a multi-scale convolutional neural network and weighted naive bayes algorithm (MCNN–WNB). We define the feature vector characterization index for Microblog negative review data and preprocess the data accordingly. We quantify the relationship between attributes and categories using the weighted Naive Bayes method and use the quantification value as the weighting coefficient for the attributes, addressing the issue of decreased classification performance in traditional methods. We introduce a sentiment classification model based on word vector representation and a multi-scale convolutional neural networks to filter out Microblog negative review data. We conduct simulation experiments using real data, analyzing key influencing parameters such as convergence time, training set sample size, and number of categories. By comparing with K-means, Naive Bayes algorithm, Spectral Clustering algorithm and Autoencoder algorithm, we validate the effectiveness of our proposed method. We discover that the convergence time of the MCNN–WNB algorithm increases as the number of categories increases. The average classification accuracy of the algorithm remains relatively stable with varying test iterations. The algorithm’s precision increases with the number of training set samples and eventually stabilizes.

Data-Driven Clustering and Feature-Based Retail Electricity Pricing with Smart Meters

The adoption of smart meters and dynamic pricing programs is rapidly increasing among electric utility companies. In “Data-Driven Clustering and Feature-Based Retail Electricity Pricing with Smart Meters,” Keskin, Li, and Sunar analyze how utility companies should use smart meter data for better pricing decisions. Utility companies typically have access to consumption patterns and high-dimensional features on customer characteristics and exogenous factors. The authors identify that such feature data can exhibit different forms of heterogeneity—over time and over customers. They show that the different forms of feature heterogeneity significantly worsen the best profit performance that can be achieved by a data-driven dynamic pricing policy. The authors also develop a policy based on joint spectral clustering and contextual dynamic pricing and prove that this policy achieves near-optimal profit performance.

Clustering-based hyper-heuristic algorithm for multi-region coverage path planning of heterogeneous UAVs

In the context of multi-heterogeneous UAV coverage path planning, an effective solution method has been proposed. Firstly, regions are set up as fully connected graphs which are cut into multiple subgraphs by spectral clustering method to assign tasks to multi-heterogeneous UAVs. Additionally, an RL-based hyper-heuristic algorithm is proposed. Heuristic space is parameterized by GNN which is trained with the reward provided by the optimization goal to automate design and enhance the heuristic metrics, avoiding the inefficiency and suboptimality of expert design and manual parameter tuning. Compared with existing methods, the proposed algorithm has a better performance in task completion time, execution time and deviation rate, which shows its potential application in the coverage path planning problem of multi-heterogeneous UAVs.

Fast adaptively balanced min-cut clustering

Min-cut clustering is a typical graph clustering method, which has been extensively applied for pattern recognition, data analysis and image processing. However, min-cut has trivial solutions, which leads to skewed clustering performance. Spectral clustering (SC) alleviates this by relaxing indicator matrix into continuous embedding and then discretizes the embedding. However, there are two primary challenges in SC, which are high computational complexity and two-stage process. To resolve these issues, a fast adaptively balanced min-cut clustering model (FBMC) is proposed, which directly solves the discrete indicator matrix without any post-processing. We utilize the bipartite graph to accelerate the construction of affinity graph and process of optimization. Besides, the balance factors are added in the model, which could alleviate skewed clustering results. Two specific methods are proposed, in which one adds a balance factor to all clusters while the other assigns balance factors to each cluster separately, referred to as FBMC1 and FBMC2. What is more, a one-step optimization is proposed to solve FBMC1 and FBMC2, the complexity of which is linear of time. Finally, comprehensive experiments results highlight the superior performance of our proposed method.

Flexible and Parameter-Free Graph Learning for Multi-View Spectral Clustering

Flexible and Parameter-Free Graph Learning for Multi-View Spectral Clustering

Repatterning of mammalian backbone regionalization in cetaceans

Cetacean reinvasion of the aquatic realm is an iconic ecological transition that led to drastic modifications of the mammalian body plan, especially in the axial skeleton. Relative to the vertebral column of other mammals that is subdivided into numerous anatomical regions, regional boundaries of the cetacean backbone appear obscured. Whether the traditional mammalian regions are present in cetaceans but hard to detect due to anatomical homogenization or if regions have been entirely repatterned remains unresolved. Here we combine a segmented linear regression approach with spectral clustering to quantitatively investigate the number, position, and homology of vertebral regions across 62 species from all major cetacean clades. We propose the Nested Regions hypothesis under which the cetacean backbone is composed of six homologous modules subdivided into six to nine post-cervical regions, with the degree of regionalization dependent on vertebral count and ecology. Compared to terrestrial mammals, the cetacean backbone is less regionalized in the precaudal segment but more regionalized in the caudal segment, indicating repatterning of the vertebral column associated with the transition from limb-powered to axial-driven locomotion.

Advanced neurological activity status of athletes based on big data technology

Currently, the application scope of big data (BD for short here) technology is relatively narrow, mostly used in the medical field, and the degree of application is relatively superficial, mostly for data statistical record analysis. Therefore, By combining the literature review, this article has decided to construct a system based on BD technology for analyzing the advanced neural activity status of athletes. The system is mainly divided into two parts, one is the biological information collection part. As an important source of system data, it is necessary to use professional equipment to collect ECG and EEG data and ensure the accuracy of the data through signal filtering, Gaussian noise elimination, salt and pepper noise, and exponential noise de-noising technology. The other is the algorithm problem of BD systems. Considering that the traditional algorithm can not deal with a large amount of data effectively, this paper chooses the BD spectral clustering algorithm based on core points as the main algorithm to cluster the data. By evaluating the efficiency of system learning, data collection and classification, system scheme construction, and error rate, this article ultimately determined the practical feasibility and effectiveness of the system. After completing the construction of the system, considering the gap between the system's performance and traditional data, this article analyzed the improvement data of various aspects of sports training. This paper compares the performance differences between the system based on BD technology and the traditional data analysis method under different indicators. In terms of data collection and classification, the accuracy of the system based on BD technology in the collection and classification of ECG and EEG data reached 100 % and 90 %, respectively, which was significantly higher than 60 % and 30 % of the traditional methods. By comparing the data from five training courses, it is found that the training efficiency of the conventional method has increased by 60 % in the first course, while the efficiency of the training method based on the BD system has increased by 85 % in the fifth course. For the activation efficiency of nerve function, the activation efficiency of brain nerve function reached 60 % and 90 % respectively in the two nerve function activation training based on the BD system, which was much higher than 30 % and 45 % of the traditional methods. Through a series of tests and comparative analysis of data, the effectiveness of the BD system is finally determined, which can achieve the goal of improving athletes' training efficiency.

Unsupervised selective labeling for semi-supervised industrial defect detection

In industrial detection scenarios, achieving high accuracy typically relies on extensive labeled datasets, which are costly and time-consuming. This has motivated a shift towards semi-supervised learning (SSL), which leverages labeled and unlabeled data to improve learning efficiency and reduce annotation costs. This work proposes the unsupervised spectral clustering labeling (USCL) method to optimize SSL for industrial challenges like defect variability, rarity, and complex distributions. Integral to USCL, we employ the multi-task fusion self-supervised learning (MTSL) method to extract robust feature representations through multiple self-supervised tasks. Additionally, we introduce the Enhanced Spectral Clustering (ESC) method and a dynamic selecting function (DSF). ESC effectively integrates both local and global similarity matrices, improving clustering accuracy. The DSF maximally selects the most valuable instances for labeling, significantly enhancing the representativeness and diversity of the labeled data. USCL consistently improves various SSL methods compared to traditional instance selection methods. For example, it boosts Efficient Teacher by 5%, 6.6%, and 7.8% in mean Average Precision(mAP) on the Automotive Sealing Rings Defect Dataset, the Metallic Surface Defect Dataset, and the Printed Circuit Boards (PCB) Defect Dataset with 10% labeled data. Our work sets a new benchmark for SSL in industrial settings.

Using Machine Learning and Natural Language Processing for Unveiling Similarities between Microbial Data

Microbiota analysis can provide valuable insights in various fields, including diet and nutrition, understanding health and disease, and in environmental contexts, such as understanding the role of microorganisms in different ecosystems. Based on the results, we can provide targeted therapies, personalized medicine, or detect environmental contaminants. In our research, we examined the gut microbiota of 16 animal taxa, including humans, as well as the microbiota of cattle and pig manure, where we focused on 16S rRNA V3-V4 hypervariable regions. Analyzing these regions is common in microbiome studies but can be challenging since the results are high-dimensional. Thus, we utilized machine learning techniques and demonstrated their applicability in processing microbial sequence data. Moreover, we showed that techniques commonly employed in natural language processing can be adapted for analyzing microbial text vectors. We obtained the latter through frequency analyses and utilized the proposed hierarchical clustering method over them. All steps in this study were gathered in a proposed microbial sequence data processing pipeline. The results demonstrate that we not only found similarities between samples but also sorted groups’ samples into semantically related clusters. We also tested our method against other known algorithms like the Kmeans and Spectral Clustering algorithms using clustering evaluation metrics. The results demonstrate the superiority of the proposed method over them. Moreover, the proposed microbial sequence data pipeline can be utilized for different types of microbiota, such as oral, gut, and skin, demonstrating its reusability and robustness.

Identification of Multi-Microgrid Clusters Using Terminal Spectral Clustering Algorithm

In response to the increasing impact of extreme weather on power distribution networks (PDNs), prioritizing resilience is imperative. This study introduces an innovative k-means spectral clustering algorithm to define the boundaries of microgrids (MGs) within a multi-microgrid (MMG) system. The aim is to improve reliability by clustering PDNs into resilient MGs. The power systems are modeled with nodes representing buses, and connections are represented as edges. The analysis involves computing the adjacency matrix, degree matrix, Laplacian matrix, and applying k-means clustering to group buses based on terminal point features. Silhouette coefficients (SC) are calculated to assess the quality of the clustering. The proposed method is tested on three IEEE distribution systems: IEEE 33, 69, and 118 bus systems. Findings reveal distinct clusters within each system with SC values above 0.68, particularly emphasizing the significance of terminal points as the basis for assisting power engineers in decision-making for predetermined grid partitioning.

An improved spectral clustering method for accurate detection of brain resting-state networks

This paper proposes a data-driven analysis method to accurately partition large-scale resting-state functional brain networks from fMRI data. The method is based on a spectral clustering algorithm and combines eigenvector direction selection with Pearson correlation clustering in the spectral space. The method is an improvement on available spectral clustering methods, capable of robustly identifying active brain networks consistent with those from model-driven methods at different noise levels, even at the noise level of real fMRI data.

Dual subspace clustering for spectral-spatial hyperspectral image clustering

Subspace clustering supposes that hyperspectral image (HSI) pixels lie in a union vector spaces of multiple sample subspaces without considering their dual space, i.e., spectral space. In this article, we propose a promising dual subspace clustering (DualSC) for improving spectral-spatial HSIs clustering by relaxing subspace clustering. To this end, DualSC simultaneously optimizes row and column subspace-representations of HSI superpixels to capture the intrinsic connection between spectral and spatial information. From the new perspective, the original subspace clustering can be treated as a special case of DualSC that has larger solution space, so tends to finding better sample representation matrix for applying spectral clustering. Besides, we provide theoretical proofs that show the proposed method relaxes the subspace space clustering with dual subspace, and can recover subspace-sparse representation of HSI samples. To the best of our knowledge, this work could be one of the first dual clustering method leveraging sample and spectral subspaces simultaneously. As a result, we conduct several clustering experiments on four canonical data sets, implying that our proposed method with strong interpretability reaches comparable performance and computing efficiency with other state-of-the-art methods.

From Affect Regulation to BMI: Unveiling Childhood Obesity Patterns Through Advanced Clustering Techniques.

This study explores the relationship between psychological factors and children's BMI, using clustering methods like Gaussian Mixture Models and Spectral Clustering. Affinity Propagation was particularly effective, suggesting that tailored interventions based on psychological assessments could improve obesity management in children.

A reconstruction-based secondary decomposition-ensemble framework for wind power forecasting

Accurate wind power forecasting remains challenging due to the instability and volatility of wind power generation. Decomposition methods are widely used to improve forecasting performance by extracting complex fluctuation patterns from wind power series. However, previous decomposition-based models ignore the global interactions across sub-signals when forecasting the sub-signals separately and miss critical local details in sub-signals when modelling all the sub-signals in a single forecasting model. To address this issue, we propose a novel “Reconstruction-based Secondary Decomposition-Ensemble (RSDE)” framework for wind power forecasting, which simultaneously preserves the global interactions and local details. Firstly, an RSD method is adopted to extract fluctuation patterns from different frequency domains: decomposing the wind power by complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), reconstructing the primary sub-signals from specific frequency domains by spectral clustering (SC) and further decomposing the reconstructed sub-signals by singular spectrum analysis (SSA). Secondly, temporal convolutional networks (TCNs) are applied to forecast each reconstructed sub-signal by fitting the corresponding secondary decomposed sub-signals, which could effectively capture the local and global features from specific frequency domains. Finally, an ensemble strategy with error correction is adopted to obtain the final forecasting results by combining the forecasted reconstructed sub-signals and corresponding error forecasting results. Four wind power datasets with different time resolutions are introduced to evaluate the forecasting performance. The experimental results demonstrate that the proposed RSDE framework consistently outperforms the benchmark models. Moreover, the proposed sub-signal modelling strategy improves the forecasting performance by more than 30% on average, and the ensemble strategy with error correction improves the forecasting performance by about 10% on average.

Examining physical activity clustering using machine learning revealed a diversity of 24-hour step-counting patterns

BackgroundPhysical activity is a crucial aspect of health benefits in the public society. Although studies on the temporal physical activity patterns might lead to the protocol for efficient intervention/program, a standardized procedure to determine and analyze the temporal physical activity patterns remains to be developed. Here, we attempted to develop a procedure to cluster 24-hour patterns of physical activity as step counts measured with an accelerometer-based wearable sensor.MethodsThe 1 Hz step count data was collected by a hip-worn triaxial accelerometer from 42 healthy participants, comprising 35 males and 7 females, at the Sendai Oroshisho center in 2008. This is a cross sectional study using unsupervised machine learning, specifically the kernel k-means algorithm with the global alignment kernel was applied on a total of 815 days from 42 participants, and 6 activity patterns were identified. Further, the probability of each 24-hour step-counting pattern was calculated for every participant and used for spectral clustering of step-behavioral patterns.ResultsWe could identify six 24-hour step-counting patterns and five daily step-behavioral clusters. We could further identify five step-behavioral clusters, all-day dominant (21 participants), all-day + bi-phasic dominant (8 participants), bi-phasic dominant (6 participants), all-day + evening dominant (4 participants), and morning dominant (3 participants). When the amount of physical activity was categorized into tertile groups reflecting highly active, moderately active, and low active, each tertile group consisted of different proportions of six 24-hour step-counting patternsConclusionsOur study introduces a novel approach using an unsupervised machine learning method to categorize daily hourly activity, revealing six distinct step counting patterns and five clusters representing daily step behaviors. Our procedure would be reliable for finding and clustering physical activity patterns/behaviors and reveal diversity in the categorization by a traditional tertile procedure using total step amount.

Research on the Implementation of Financial Risk Control Models and Algorithms Based on Machine Learning

Nowadays, machine learning has been rapidly applied to the financial services industry. With the applications of machine learning, it will bring challenges to the stable development of the financial industry. This paper studies the user behavior analysis algorithm based on operator big data, and provides financial risk control mechanism for financial enterprises. Taking finance as the data collection object, in the user behavior prediction algorithm, the experiment uses the spectral clustering algorithm to cluster the non-convex data samples, and converts the samples to the Laplacian feature space. It not only reduces the dimension, but also clusters the data, and obtains the global optimal solution. The proposed algorithm is analyzed experimentally, and the tool prototype is realized. The developed gadgets have also been applied in enterprises. The research provides a method for the construction of a risk control model in the financial industry, thereby helping financial enterprises to better control the risk of lending. This paper also provides reference for the performance improvement of the model.

Electrokinetics with Surface-Enhanced Raman Spectroscopy for Bacterial Identification in Wastewater

Bacterial presence in wastewater serves as a crucial bioindicator for monitoring population-level infections and offering early disease outbreak warnings [1,2]. However, identifying bacteria in wastewater presents outstanding challenges; notably, current culturing or fluorescence-based methods [3] to identify bacteria are slow and expensive, unsuitable for high-throughput screening of diverse bacterial species, and may not work well in the complex wastewater matrix. Here, we harness the potential of surface chemistry and electrokinetics to introduce an innovative electro-optical approach that has the promise to detect a broad spectrum of pathogenic bacteria in wastewater.We combine surface-enhanced Raman spectroscopy (SERS) [4] with alternating current electric fields, complemented by machine learning models [5], to enable rapid and amplification-free bacteria detection in filtered wastewater, at cell concentrations as low as 104 cells/mL. First, we synthesize gold nanorods that can electrostatically bind to bacteria surfaces, allowing for SERS from cell surfaces. We collect SERS from bacteria spiked into filtered wastewater, including Staphylococcus aureus, Staphylococcus epidermidis, and Escherichia coli, spanning concentrations from 109 cells/mL to 104 cells/mL. Spectral clustering analysis shows that bacteria signals become more indistinguishable from wastewater as the concentrations decrease. Further, we incorporate electrokinetic effects into SERS by employing gold microelectrodes to apply electric fields, utilizing dielectrophoresis [6] to rapidly displace and concentrate bacteria based on their dielectric responses within minutes. The enrichment of bacteria with nanorods is directly visualized through microscopy, resulting in Raman signal intensities increased by up to tenfold under electrical fields for bacterial concentrations ranging from 106 down to 104 cells/mL. Such enhancement may enable the detection sensitivity to reach environmentally-relevant concentrations. Moreover, employing data science analysis, we identify biologically-relevant “fingerprint” Raman peaks characterizing proteins, nucleic acids, and lipids from bacteria surfaces, allowing for rapid identification of bacteria species in wastewater. This novel method exhibits potential for generalized pathogen detection and molecular recognition in complex liquid samples, such as wastewater, blood, and seawater.[1] Hellmér, et al. Applied and environmental microbiology. (2014).[2] Keshaviah, et al. The Lancet Global Health. (2023).[3] Jahn, et al. Nature Microbiology. (2022).[4] Tadesse, et al., Nano Lett. (2020).[5] Ho, et al., Nat. Comm. (2019).[6] Pethig. John Wiley & Sons (2010).

Spectral ensemble clustering with doubly stochastic co-association matrix

Although co-association (CA) matrix-based spectral ensemble clustering (SEC) has achieved significant success, it faces two persistent challenges: (1) CA matrix-based SEC may result in imbalanced cluster outcomes, and (2) SEC encounters difficulties due to the high time and space complexity resulting from the quadratic scaling of the CA matrix with the number of data points. To tackle these issues, we propose a doubly stochastic co-association (DSCA) matrix construction method. We have theoretically demonstrated that DSCA matrix-based SEC can achieve a more balanced cluster outcome compared to CA matrix-based SEC. Leveraging the DSCA matrix, we introduce two novel methods: Fast SEC (FSEC) and Approximate SEC (ASEC). FSEC efficiently solves the normalized cut on the DSCA matrix as it is symmetric and doubly stochastic. In ASEC, we employ random samples rather than the entire big dataset for ensemble clustering. After obtaining the ensemble clustering result of benchmark random sample, the approximate ensemble clustering result for all data points is derived using our newly proposed probability nearest neighbors (PNN) algorithm. Experimental results on both real-world and synthetic datasets are evaluated in terms of accuracy (ACC), normalized mutual information (NMI), purity, F1-score, and adjusted rand index (ARI) and confirm the scalability and effectiveness of the proposed FSEC and ASEC.