Data Clustering Research Articles

Interactive Geographic Visualization and Unsupervised Learning for Optimal Assignment of Preachers to Appropriate Congregations

Riau Province has a population of 6,642,874 and a diverse geography, which poses significant challenges in optimizing Islamic preaching activities. Traditional assignment methods often lead to inefficiencies due to misalignment between the preacher’s expertise and congregational needs, as well as logistical issues. This study integrates K-Means clustering and DBSCAN algorithms with interactive geographic visualization to optimize the assignment of preachers to mosques. We collected 435 data points, including 185 mosques and 250 preachers. K-Means was evaluated using the Elbow Method and Silhouette Score, identifying 10 clusters as optimal with a Silhouette Score of 0.435654. However, K-Means does not handle outliers effectively, as indicated by zero outliers in all configurations. DBSCAN was tested with various epsilon (eps) and minimum sample values. The optimal configuration with eps of 1.5 and 5 minimum samples resulted in 10 clusters with a Silhouette Score of 0.381108 and 60 outliers. DBSCAN effectively manages outliers and varying densities. Although K-Means is advantageous for its simplicity and higher Silhouette Scores, it is unable to handle outliers effectively. DBSCAN provides robust clustering for noisy data. Therefore, it can be concluded that hybridizing unsupervised learning algorithms with geographic visualization can potentially improve the effectiveness of preaching activities in Riau Province and enhance preacher assignment.

Organization processes and artificial intelligence (AI) for healthcare processes reorganization: a case study

PurposeThis study aims to provide a methodology and tools to design new organizational processes and artificial intelligence (AI)-based scoring to optimize the resources management in healthcare units.Design/methodology/approachProcess design and process data-driven simulation: the processes are designed by the business process modeling and notation and the unified modeling language standards. Data processing is performed by Correlation matrix analysis and by Fuzzy c-Means data clustering. The matching between the two methods provides the most indicated final corrective actions of the “TO BE” organizational model.FindingsThis proposed method, experimentally applied in this work merging the lean management model (LMM), process mining (PM) and AI methods, named process mining organization (PMO) model (Rosa et al., 2023 (b)), is able to improve organizational processes of a hospitalization unit (HU) by developing three propaedeutic phases: (1) analysis of the current state of the processes (“AS IS”) by identifying the critical issues as bottlenecks of processes, (2) AI data processing able to provide additional classified and predicted information allowing the “TO BE” workflow process and (3) implementation of corrective actions suggested by the PMO in order to support strategic decision-making processes in the short, medium and long term by classifying an order of priority about the healthcare procedures/protocols to perform.Research limitations/implicationsThe main limitation of the proposed case study is in the limited number of available digital data to process. This aspect reduces the capability to interpret result. In any case, the proposed methodology is a “launch” work to define a new approach to integrate organizational processes including workflow design and AI scoring. Future work will be focused on managerial implications due to use of the discussed method: design and development of new human resource (HR) organizational protocols following data analysis to optimize costs and care services and to decrease injury compensation claims.Practical implicationsMain implications are in healthcare managerial scenarios: design and development of new HR organizational protocols following data analysis to optimize costs and care services and to decrease injury compensation claims.Social implicationsCare services optimization is addressed on HUs.Originality/valueThe design of HR organizational processes integrates AI-driven data decision-making processes. This case study examines AI-based innovation analytics addressed on resource efficiency.

A Unified Graph Theory Approach: Clustering and Learning in Criminal Data

Crime report clustering plays a critical role in modern law enforcement, enabling the identification of patterns and trends essential for proactive policing. However, traditional clustering approaches face significant challenges with the complex, unstructured nature of crime reports and their inherent sparse relationships. While graph-based clustering shows promise, issues of noise sensitivity and data sparsity persist. This study introduces a unified approach integrating spectral graph-based clustering with Graph Convolutional Networks (GCN) to address these challenges. The proposed approach encompasses data collection, preprocessing, linguistic feature extraction, vectorization, graph construction, graph learning, and clustering to effectively capture the intricate similarities between crime reports. The proposed approach achieved significant improvements over existing methods: a Silhouette Score of 0.77, a Davies–Bouldin Index of 0.51, and consistent performance across varying dataset sizes (100–1000 nodes). These results demonstrate the potential for enhanced crime pattern detection in law enforcement operations.

Early Fault Diagnosis and Prediction of Marine Large-Capacity Batteries Based on Real Data

The inconsistency of battery voltages in all-electric ships is a significant issue for electric vehicle battery systems, leading to numerous safety concerns during vessel operation. Therefore, timely fault diagnosis and accurate fault prediction are crucial for the safe operation of ships. This study examines the fault alarm system of marine battery management systems in conjunction with the unique operating conditions of ships, focusing on the system’s latency. To facilitate prompt fault detection, a fault diagnosis method based on the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is proposed, utilizing the voltage data of battery clusters. Results indicate that the DBSCAN clustering algorithm demonstrates superior effectiveness and accuracy in identifying irregular battery clusters. Furthermore, the fault prediction method based on the iTransformer model is introduced to forecast variations in battery cluster voltages. Experimental findings suggest that this model can effectively predict consistency faults and over-/under-voltage conditions based on battery cluster voltage values and corresponding fault thresholds.

Research on NBA Event Prediction Based on Hybrid Machine Learning Model

NBA is one of the top basketball events in the world. It is of great significance to predict NBA events comprehensively and accurately. Therefore, this paper puts forward the research on the prediction of NBA events based on a hybrid machine learning model. Construct a hybrid machine learning model, set the pre-processing flow of NBA event prediction data, solve the probability core cluster of NBA event prediction data, and form a data stream mining matrix of NBA event prediction; based on the integrated learning method in machine learning, according to the principle of constructing the prediction index of NBA events, the prediction index is constructed; based on the deep learning method in machine learning, the weight of target factors is used as a parameter to measure the prediction accuracy, and hierarchical evaluation is carried out to realize the prediction of NBA events. The experimental results show that the research method shows a high goodness of fit in all NBA games, and the predicted results are highly consistent with the actual results, and the predicted results are closer to the real results, with small root mean square error and average absolute error, which has a good prediction effect.

Unified representation aggregated from graph and global features for multi-source data clustering

Multi-source data has received increasing attention due to its excellent performance in clustering tasks. However, existing multi-source data clustering methods utilize shallow graph learning methods to model similarity graphs of multi-source data, and ignore the importance of weak connections between samples within the same cluster. Thus, these connections fail to be explored in the global similarity graph of multi-source data. In this paper, we proposed a unified representation aggregated from graph and global features (UGGF) method for multi-source data clustering. Specifically, it utilizes the cross-source graph diffusion process to preserve invariant connections in multi-source similarity graphs and preserve weak connections between samples through their connection relationships between different sources. Furthermore, inspired by self-attention, the encoder of transformers is used to learn multi-source global feature representation. Then, global features and similarity graphs are integrated to comprehensively explore the multi-source data, aiming to obtain a unified representation for the final clustering task. The proposed UGGF method is validated on four benchmark datasets. Extensive experimental results demonstrate the superiority of the proposed method compared with other state-of-the-art multi-source data clustering methods.

Leveraging Transactional Data for Analytics: Building Economic Indicators from Wire Transfers, Credit Card, and ACH Transactions

Traditional financial metrics for startup companies are often scarce due to the limited availability of public financial data. This paper proposes a novel approach to building leading economic indicators for startups by leveraging transactional data from wire transfers, credit card transactions, and Automated Clearing House (ACH) payments. These data sources provide rich, granular insights into the financial activities of startups, offering a proxy for their operational health and growth trajectories. The methodology involves clustering transactional data based on sender and receiver information to identify spending patterns. Outgoing transactions are categorized into key expenditure categories such as payroll, marketing, legal, computing, and operational costs. By analyzing the distribution and frequency of these expenses, this framework provides a comprehensive view of a startup's resource allocation and financial priorities. In addition, incoming transactions from various channels are used as a proxy for sales or revenue, enabling the development of dynamic indicators of financial performance. This approach bridges the gap in financial visibility for startups, providing timely and actionable insights into their economic activities. These leading indicators can serve as early warning signals of financial distress or as predictors of growth, offering immense value to investors, analysts, and policymakers. The study highlights the potential of transactional data analytics in enhancing transparency and decision-making in the startup ecosystem, paving the way for more robust analytical frameworks in a traditionally opaque domain. Index Terms: Transactional Data Analysis, Wire Transfers, Credit Card Transactions, ACH Payments, Startup Financial Health, Spending Patterns, Revenue Proxy, Data Clustering, Expense Categorization, Payroll Analysis, Marketing Expenses, Legal Costs, Computing and Operational Costs, Startup Growth Metrics, Financial Visibility, Early Warning Signals, Startup Ecosystem Analytics

Impact Analysis of the Digital Entrepreneurial Ecosystem to Improve the Tourism Industry and Social Sustainability

Digitalisation is a key enabler of sustainable tourism in an industry that has recently been transformed by new sustainable innovations and digital solutions. In this study, we analysed the effects of the digital entrepreneurial ecosystem (DEE) on the tourism industry and social sustainability of 27 EU countries. The study underlines the key independent indicators representing the impact of the DEE’s elements. Also, a quantitative and comparative approach was considered using the panel data method and clustering analysis for data from 2014 to 2021. Our findings show significant positive impacts of DEE elements that have significantly contributed to tourism and social sustainability growth. Furthermore, hierarchical clustering analysis (HCA) revealed that eight countries (cluster A), including Germany, the UK, France, Spain, and Italy, had the highest average digitalisation levels, affecting their tourism growth and social sustainability. Ultimately, we indicate different digital user levels and marketplace, such as networks designed to produce cloud infrastructure, digital platforms, and digital tourist-based devices and applications, as having the capability to enhance tourism and social sustainability.

Cybersecurity of smart grids: Comparison of machine learning approaches training for anomaly detection

Objectives. The transformation of modern electric grids into decentralized smart grids presents new challenges in the field of cybersecurity. The purpose of this work is to conduct research and analysis into the effectiveness of different machine-learning methods for identifying anomalies in decentralized smart networks, including cyberattacks and emergency modes, as well as to develop recommendations on the optimal combination of these methods for ensuring effective cybersecurity under conditions of changing electrical loads.Methods. We consider several machine learning methods for identifying anomalies in power systems that simulate network behavior under conditions of cyberattacks and emergency modes. The relative effectiveness of such methods as multifractal analysis using wavelets, the Isolation Forest model, local outlier factor (LOF), k-means clustering, and one-class support vector machine (One-Class SVM), is analyzed.Results. The comparison of machine learning methods reveals the varying effectiveness of anomaly detection methods used to detect cyber threats and deviations in electrical systems. Isolation Forest is best at detecting abrupt changes related to cyberattacks with high accuracy and a minimum of false positives. While LOF can also be effective in detecting cyberattacks, its increased sensitivity to minor deviations increases the number of false positives. K-means and One-Class SVMs are less effective in detecting abrupt anomalies but are useful for general clustering of data and detecting both abrupt and smooth changes, respectively.Conclusions. The obtained research results indicate the advantages of using a combination of machine learning algorithms to ensure the reliable protection of smart networks from cyberattacks taking into account the nature of the electrical load.

Sztuczna inteligencja (AI) w modyfikacjach biznesowych i niezauważone przejście od informacji do inwigilacji

Artificial Intelligence (AI) is becoming the foundation of IT systems and will gradually take over many decision-making processes. Passion for modern technologies is commonly combined with many years of experience gained in analyzing large data sets and creating analytical applications. Interdisciplinary teams specializing in the use of artificial intelligence and machine learning tools (Machine Learning), in particular classification and prediction models and data clustering, are formed. In the new development conditions, there were radical restrictions on social and economic activity and the emergence of new barriers, which resulted in an increase in the importance of innovative products and services based on modern technologies, aimed at overcoming the economic and social effects of the COVID-19 pandemic. Trends such as gamification, Big Data and artificial intelligence (AI) were visible. Thanks to this, the right vision was created, supported by the necessary technical knowledge to implement solutions that open the gate of undiscovered potential. Acceleration of digitization processes resulted in long-term organizational changes, including more frequent remote work of some employees. Uncertainty in the global economy caused by the pandemic has prom pated many companies reassess their development models. Instead of relying on global supply chains, more and more companies have invested in robots, resulting in a manufacturing renaissance in industrialized countries. The main aim of the study is to present artificial intelligence and the threats of its use.

Genetic Identity and Diversity of Loggerhead Sea Turtles in the Central Mediterranean Sea

Background: The conservation of loggerhead sea turtles (Caretta caretta) in the central Mediterranean benefits from an in-depth understanding of its population genetic structure and diversity. Methods: This study, therefore, investigates C. caretta in Maltese waters by genetically analysing 63 specimens collected through strandings and in-water sampling, using mitochondrial DNA control region and microsatellites. Additionally, the two nests detected in Malta in 2023 were analysed for the same markers. Results: Mitochondrial data identified 10 haplotypes, with mixed stock analyses tracing 87.5% of the specimens to Mediterranean origins, primarily from Libyan rookeries, with contributions from Lebanon, Israel and Turkey. Three Atlantic haplotypes were identified in six specimens, with CC-A17.1 linking central Mediterranean foraging individuals to rookeries in Cape Verde. Five of these six Atlantic haplotype records were from recently sampled individuals (2022–2023), possibly indicating a recent eastward expansion of Atlantic haplotypes into the Mediterranean. Bayesian clustering (K = 2) of microsatellite data using haplotypes as priori revealed similar proportions for clusters across most specimens, except for three specimens with Atlantic haplotypes CC-A1.1 and CC-A1.3, which exhibited distinct patterns. The two nests examined here displayed Mediterranean haplotypes, with nuclear DNA matching the predominant Mediterranean profiles found in foraging individuals, suggesting that local clutches originated from Mediterranean parents. Conclusions: Increasing nesting activity on Maltese beaches and this archipelago’s geographical position highlight the need for ongoing genetic monitoring to track changes in genetic diversity and develop conservation strategies that support the effective protection of this species and its habitats.

Evaluation of grounding grid corrosion extent based on laser-induced breakdown spectroscopy (LIBS) combined with machine learning

As one of the most widely used forms of energy, the safety and stability of power systems are crucial to modern society. Grounding grids dissipate current and reduce touch and pace voltage during lightning strikes or fault currents, ensuring the safety of personnel and equipment. However, prolonged submersion in soil causes inevitable corrosion, compromising grounding efficacy by increasing resistance and reducing current dissipation. This deterioration can result in unsafe local potential differences. This study uses Laser-Induced Breakdown Spectroscopy (LIBS) to measure corrosion degrees in grounding grids. Spectral data from samples with varying corrosion extent were collected, with outliers removed using the Local Outlier Factor (LOF) algorithm. Principal Component Analysis (PCA) reduced data dimensionality, revealing clustering in spectral data corresponding to corrosion extent. Three machine learning models were compared: Adaptive Boosting - Backpropagation Neural Network (Adaboost-BP), Support Vector Machine (SVM), and Random Forest (RF). The RF model showed the highest accuracy in predicting corrosion degree (R²=0.9845, MSE=0.0296), outperforming Adaboost-BP and SVM, especially for intermediate corrosion extent. These findings validate the effectiveness and reliability of combining LIBS with machine learning for predicting grounding grid corrosion, providing a theoretical foundation for the safe operation of power systems.

ScQA: A dual-perspective cell type identification model for single cell transcriptome data

Single-cell RNA sequencing technologies have been pivotal in advancing the development of algorithms for clustering heterogeneous cell populations. Existing methods for utilizing scRNA-seq data to identify cell types tend to neglect the beneficial impact of dropout events and perform clustering focusing solely on quantitative perspective. Here, we introduce a novel method named scQA, notable for its ability to concurrently identify cell types and cell type-specific key genes from both qualitative and quantitative perspectives. In contrast to other methods, scQA not only identifies cell types but also extracts key genes associated with these cell types, enabling bidirectional clustering for scRNA-seq data. Through an iterative process, our approach aims to minimize the number of landmarks to approximately a dozen while maximizing the inclusion of quasi-trend-preserved genes with dropouts both qualitatively and quantitatively. It then clusters cells by employing an ingenious label propagation strategy, obviating the requirement for a predetermined number of cell types. Validated on 20 publicly available scRNA-seq datasets, scQA consistently outperforms other salient tools. Furthermore, we confirm the effectiveness and potential biological significance of the identified key genes through both external and internal validation. In conclusion, scQA emerges as a valuable tool for investigating cell heterogeneity due to its distinctive fusion of qualitative and quantitative facets, along with bidirectional clustering capabilities. Furthermore, it can be seamlessly integrated into border scRNA-seq analyses. The source codes are publicly available at https://github.com/LD-Lyndee/scQA.

Deep clustering analysis via variational autoencoder with Gamma mixture latent embeddings

This article proposes a novel deep clustering model based on the variational autoencoder (VAE), named GamMM-VAE, which can learn latent representations of training data for clustering in an unsupervised manner. Most existing VAE-based deep clustering methods use the Gaussian mixture model (GMM) as a prior on the latent space. We employ a more flexible asymmetric Gamma mixture model to achieve higher quality embeddings of the data latent space. Second, since the Gamma is defined for strictly positive variables, in order to exploit the reparameterization trick of VAE, we propose a transformation method from Gaussian distribution to Gamma distribution. This method can also be considered a Gamma distribution reparameterization trick, allows gradients to be backpropagated through the sampling process in the VAE. Finally, we derive the evidence lower bound (ELBO) based on the Gamma mixture model in an effective way for the stochastic gradient variational Bayesian (SGVB) estimator to optimize the proposed model. ELBO, a variational inference objective, ensures the maximization of the approximation of the posterior distribution, while SGVB is a method used to perform efficient inference and learning in VAEs. We validate the effectiveness of our model through quantitative comparisons with other state-of-the-art deep clustering models on six benchmark datasets. Moreover, due to the generative nature of VAEs, the proposed model can generate highly realistic samples of specific classes without supervised information.

Multi-View Adaptive Fusion Network for Spatially Resolved Transcriptomics Data Clustering

Multi-View Adaptive Fusion Network for Spatially Resolved Transcriptomics Data Clustering

An AI-Based Deep Learning with K-Mean Approach for Enhancing Altitude Estimation Accuracy in Unmanned Aerial Vehicles

In the rapidly evolving domain of Unmanned Aerial Vehicles (UAVs), precise altitude estimation remains a significant challenge, particularly for lightweight UAVs. This research presents an innovative approach to enhance altitude estimation accuracy for UAVs weighing under 2 kg without cameras, utilizing advanced AI Deep Learning algorithms. The primary novelty of this study lies in its unique integration of unsupervised and supervised learning techniques. By synergistically combining K-Means Clustering with a multiple-input deep learning regression-based model (DL-KMA), we have achieved substantial improvements in altitude estimation accuracy. This methodology represents a significant advancement over conventional approaches in UAV technology. Our experimental design involved comprehensive field data collection across two distinct altitude environments, employing a high-precision Digital Laser Distance Meter as the reference standard (Class II). This rigorous approach facilitated a thorough evaluation of our model’s performance across varied terrains, ensuring robust and reliable results. The outcomes of our study are particularly noteworthy, with the model demonstrating remarkably low Mean Squared Error (MSE) values across all data clusters, ranging from 0.011 to 0.072. These results not only indicate significant improvements over traditional methods, but also establish a new benchmark in UAVs altitude estimation accuracy. A key innovation in our approach is the elimination of costly additional hardware such as Light Detection and Ranging (LiDAR), offering a cost-effective, software-based solution. This advancement has broad implications, enhancing the accessibility of advanced UAVs technology and expanding its potential applications across diverse sectors including precision agriculture, urban planning, and emergency response. This research represents a significant contribution to the integration of AI and UAVs technology, potentially unlocking new possibilities in UAVs applications. By enhancing the capabilities of lightweight UAVs, we are not merely improving a technical aspect, but revolutionizing the potential applications of UAVs across industries. Our work sets the stage for safer, more reliable, and precise UAVs operations, marking a pivotal moment in the evolution of aerial technology in an increasingly UAV-dependent world.

A Case Study on the Data Mining-Based Prediction of Students’ Performance for Effective and Sustainable E-Learning

The study explores the application of data analytics and machine learning to forecast academic outcomes, with the aim of ensuring effective and sustainable e-learning. Technological study programs in universities often experience high dropout rates, which makes it essential to analyze and predict potential risks to reduce dropout percentages. Student performance prediction (SPP) offers potential benefits, including personalized learning and early interventions. However, challenges such as (1) data quality and availability and (2) incomplete and inconsistent data complicate this process. Moreover, to support the fourth Sustainable Development Goal (SDG), we focus on the quality of education. A case study approach is used using data mining techniques, particularly classification, regression, and clustering, to predict student performance. The case presented aims to predict risks and ensure academic success and quality. The cross-industry standard process for data mining (CRISP-DM) methodology is used to structure and guide the prediction process. The study shows that using data from student learning processes within an academic success prediction model and data mining can identify at-risk students.

Application of big data and artificial intelligence in visual communication art design.

In the era of continuous development of computer technology, the application of artificial intelligence (AI) and big data is becoming more and more extensive. With the help of powerful computer and network technology, the art of visual communication (VISCOM) has ushered in a new chapter of digitalization and intelligence. How vision can better perform interdisciplinary and interdisciplinary artistic expression between art and technology and how to use more novel technology, richer forms, and more appropriate ways to express art has become a new problem in visual art creation. This essay aims to investigate and apply VISCOM art through big data and AI methods. This essay proposed the STING algorithm for big data for multi-resolution information clustering in VISCOM art. In addition, the convolutional neural network (CNN) in AI technology was used to identify the conveyed objects or scenes to achieve the purpose of designing art with different characteristics for different scenes and groups of people. STING is a multi-resolution clustering technique for big data, with the advantage of efficient data processing. In the experimental part, this essay selected a variety of design contents in VISCOM art, including logo design, text design, scene design, packaging design and poster design. STING and CNN algorithms were used to cluster and AI-identify the design elements 16 of the design projects might contain. The results showed that the overall average clustering accuracy was above 82%, the accuracy of scene element recognition mainly was above 80%, and the accuracy of facial recognition was above 80%; this showed that this essay applied AI and big data to the design of VISCOM, and had a good effect on the clustering and identification of design elements. According to expert scores, these applications' reliability and practicality scores were above 70 points, with an average of about 80 points. Therefore, applying big data and AI to VISCOM in this essay is reliable and feasible.

Fuzzy clustering of mixed data with spatial regularization

A fuzzy clustering model for data with mixed features and spatial constraints is proposed. The clustering model allows different types of variables, or attributes, to be taken into account. This result is achieved by combining the dissimilarity measures for each attribute employing a weighting scheme, to obtain a distance measure for multiple attributes. The weights are objectively computed during the optimization process. The weights reflect the relevance of each attribute type in the clustering results. A spatial term is taken into account, considering a wide definition of contiguity, either physical contiguity or the adjacency matrix in a network. Simulation studies and two empirical applications, including both physical and abstract definitions of contiguity are presented that show the effectiveness of the proposed clustering model.

Long-term traffic speed prediction utilizing data augmentation via segmented time frame clustering

Among many traffic forecasting studies, comparatively fewer studies focus on long-term traffic prediction, such as 24-hour prediction. While traffic data such as traffic speed are easier to obtain, obtaining similarly reliable and accessible feature data with the inclusion of weather or events would be difficult depending on the location or availability of the service providers. Getting these data becomes a more significant issue when considering global coverage. To mitigate the issue of limited feature data, a method to augment already existing data by improving the dataset's quality and ensuring more accurate training via sorting the dataset into appropriate clusters to be used as an additional feature is proposed. This paper proposes a long-term traffic forecasting model that utilizes a novel time-series segmentation method paired with data clustering and classification via Convolutional Neural Network (CNN) to cover the lack of traffic data and features as additional pre-processing before using Long Short-Term Memory (LSTM) for long-term traffic prediction which is not researched as much. This proposed model is called Cluster Augmented LSTM (CAL). The proposed model is compared with existing machine learning models and evaluated using Mean Absolute Percentage Error (MAPE) and Root-Mean-Squared-Error (RMSE) performance metrics. A comparison between LSTM and Gated Recurrent Units (GRU) was conducted, showing that GRU tends to outperform LSTM in most cases. However, the best-performing result for the proposed method still utilizes LSTM. The final results show that the proposed CAL model could achieve better results by 1.42 %-1.76 % and 0.25–0.41 for MAPE and RMSE, respectively.