Clustering Process Research Articles

Clustering Commuter Behavior based on Automated Fare Collection (AFC)

This paper examines the application of the Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) method to cluster Automated Fare Collection (AFC) transaction data from train travelers in Jakarta, Bogor, Depok, Tangerang, and Bekasi (Jabodetabek) in Indonesia. To enhance the clustering process, the Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction and the DenseClus library are employed. In this study, different combinations of hyperparameters are used to identify the optimal configuration for producing distinct clusters with a high concentration and noticeable distinction. The results demonstrate that the utilization of HDBSCAN on UMAP-reduced data effectively, discerning unique trip patterns and emphasizing notable disparities in travel distance, time, and length among various clusters. The UMAP intersection method showed notable efficacy in maintaining the local structure of the data, resulting in the development of distinct and meaningful clusters. In addition, categorical data were transformed into numerical formats using hashing techniques, efficiently tackling the difficulties posed by a high number of categories and assuring efficient data processing. The results reveal vital insights into the application of density-based clustering to intricate transportation data, with major implications for enhancing route planning and capacity management for Jabodetabek commuters.

Multi-Step Assembly of an RNA-Liposome Nanoparticle Formulation Revealed by Real-Time, Single-Particle Quantitative Imaging.

Self-assembly plays a critical role in nanoparticle-based applications. However, it remains challenging to monitor the self-assembly of multi-component nanomaterials at a single-particle level, in real-time, with high throughput, and in a model-independent manner. Here, multi-color fluorescence microscopy is applied to track the assembly of both liposomes and mRNA simultaneously in clinical mRNA-based cancer immunotherapy. Imaging reveals that the assembly occurs in discrete steps: initially, RNA adsorbs onto the liposomes; then, the RNA-coated liposomes cluster into heterogeneous structures ranging from sub-micrometer to tens of micrometers. The clustering process is consistent with a Smoluchowski model with a Brownian diffusion kernel. The transition between the two steps of assembly is determined by the orientation of RNA-mediated interactions. Given the facile application of this approach and the ubiquity of the components studied, the imaging and analysis in this work are readily applied to monitor multi-component assembly of diverse nanomaterials.

Application of K-Means and Naïve Bayes Algorithms for Prediction Model of Student Interest Concentration (Case Study: Amikom University Yogyakarta)

Amikom University Yogyakarta, has a Master of Informatics Engineering study program with three concentrations of specialization: Business Intelligence, Digital Information Intelligence, and Intelligence Animation. The choice of concentration by prospective students has been based on subjectivity, not on competence or work experience. To overcome this, this research proposes an algorithm-based concentration prediction and recommendation model to help prospective students choose the appropriate concentration. The dataset is obtained through questionnaires collected from active and inactive students. This research uses the K-Means algorithm for clustering raw data (unsupervised) in order to generate target classes, which are then classified using Naïve Bayes. The clustering process determines concentration labels such as Business Intelligence and others, while the SMOTE technique is used to balance the dataset to avoid data imbalance problems. This approach aims to produce more objective and accurate recommendations in determining student concentrations, reducing the tendency of subjectivity, and increasing the relevance of student competencies to the chosen field of specialization. From this research, the K-Means DBI score is 0.277 and the Naïve Bayes prediction accuracy score is 89%. This research aims to produce more objective and accurate recommendations in determining student concentrations, reducing subjectivity, and increasing the relevance of student competencies to the chosen field of specialization. The proposed model is expected to help universities in designing a more targeted admission strategy, as well as supporting students in making academic decisions that are in accordance with their abilities and interests, thereby increasing the effectiveness of the learning process and the suitability of graduates to the needs of the world of work.

On the structure of nanoparticle clusters: effects of long-range interactions.

The fractal structure of aggregates consisting of primary nanoparticles naturally arises during their synthesis. While typically considered to be a fully stochastic process, we suspect long-range interactions, in particular van der Waals forces, to induce an active pull on particles, altering the clustering process. Using an off-grid 3D model, we show that an active pull decreases the density and fractal dimension of formed clusters. These findings could not be reproduced by 2D models, which underestimate screening effects. Additionally, we determined the range within which van der Waals forces dominate the aggregation process.

MutaSwarmClus: enhancing data clustering efficiency with mutation-enhanced swarm algorithm

Data clustering is a fundamental technique in data mining, pivotal for various applications such as statistical analysis and data compression. Traditional clustering algorithms often struggle with noisy or high-dimensional datasets, hindering their efficacy in addressing real-world challenges. In response, this research introduces MutaSwarmClus, a novel hybrid metaheuristic algorithm that combines Mouth Brooding Fish (MBF), Ant Colony Optimization (ACO), and mutation operators to enhance clustering quality. MutaSwarmClus intends to adaptively control the exploration and exploitation phases of the solution space, solve issues with local optima and changes in the distribution of available data. Moreover, it incorporates an Iterated Local Search (ILS) to refine solutions and avoiding getting stuck in local optima. MutaSwarmClus therefore increases the robustness of the clustering process by incorporating controlled randomness through mutation operators in order to handle noisy and outlier data points well. According to the contributions analysis, the proposed algorithm improves the clustering solution with the combined system of MBF, ACO, and mutation operators, which enables the mechanism of exploration and exploitation in the process of information search. As shown through the results of experimental studies, MutaSwarmClus has high performance when used with various benchmarks, and outperforms or performs as well as or better than compared to other clustering algorithms such as K-means, ALO, Hybrid ALO, and MBF. It achieves an average error rate of only 10%, underscoring its accuracy in clustering tasks. The utilization of MutaSwarmClus offers a solution to the existing problems in clustering large datasets in terms of scalability, efficiency and accuracy. Possible directions for future work can continue to optimize the model parameters of the algorithm and study its adaptability in dynamic conditions and with large amounts of data.

Rancang Bangun Sistem TOEFL Untuk Analisis Kelemahan Peserta Dengan Penerapan Algoritma K-Means Clustering

English proficiency is a critical skill across various sectors, particularly in educational and professional contexts. The TOEFL test assesses English language skills, covering aspects such as Reading, Listening, Speaking, and Writing. However, TOEFL results often only provide a total score without detailed insights into participants' weaknesses. To address this, this study utilizes the K-Means Clustering algorithm for its simplicity, efficiency in processing multidimensional data, and ability to produce statistically meaningful groupings. The research aims to identify the weaknesses of TOEFL participants at the Language Laboratory of ITN Malang. The dataset comprises scores from 520 students across Reading, Structure and Written Expression, and Listening sections. This method classifies participants into three primary clusters: Reading Boosters (C1), Grammar Builders (C2), and Listening Improvers (C3). The clustering process involves randomly selecting initial centroids, calculating distances using the Euclidean Distance formula, and iterating until cluster stability is achieved. The analysis results show that 36.54% of participants fall into C1, 41.35% into C2, and 22.21% into C3. The implementation of this algorithm offers valuable insights for developing more effective learning programs.

Ion-molecule reactions in CO2 clusters induced by low-energy electron attachment

Abstract Dissociative electron attachment (DEA) process in molecular clusters is inherently complex, and elucidating the underlying mechanisms presents significant challenges. This study builds upon our previous research by investigating DEA in the larger and more numerous CO2 clusters, and extending the electron attachment energy to 15 eV. This enhanced approach allows for the detection of anionic species such as O 2 − and (CO2)nO 2 − . We recorded the production efficiency curves and kinetic energy distributions of the fragment anions, revealing that the O 2 − anions can be formed via the DEA process of CO2 monomers, consistent with previously reported observations. Additionally, O 2 − may also be produced through a cascade reaction process involving ion-molecule/atom interactions within the clusters, even when the attachment energy exceeds the DEA threshold of O 2 − production from CO2 monomers. Furthermore, the stabilization of these detected anions (CO2)nO 2 − is attributed to energy dissipation via molecular evaporative cooling, as previously proposed. This study offers new insights into the underlying mechanisms of DEA in molecular clusters, thereby enhancing our understanding of electron-induced reactions in atmospheric and astrochemical contexts.

CLUSTERING DISTRICTS/CITIES IN EAST JAVA PROVINCE BASED ON HIV CASES USING K-MEANS, AGNES, AND ENSEMBLE

Cluster analysis is a method of grouping data into certain groups based on similar characteristics. This research aims to group districts/cities in East Java Province in 2021 based on HIV cases using hierarchical cluster analysis (AGNES), non-hierarchical cluster analysis (K-means), and ensemble clustering. The study found that the ensemble clustering solution forms four clusters, consistent with the results of AGNES clustering. This suggests that ensemble clustering improves the quality of cluster solutions by leveraging both hierarchical and non-hierarchical methods. The grouping of districts/cities based on HIV cases provides a clear distribution pattern for more targeted interventions. The study is limited to HIV cases in East Java Province and may not be generalizable to other regions with different epidemic characteristics. Additionally, the study focuses on clustering methods without investigating temporal changes in HIV case distribution. This research is one of the few studies that applies ensemble clustering to HIV cases in East Java Province. It combines hierarchical and non-hierarchical methods to improve the clustering process and provides a practical approach for regional HIV control planning.

Multi-view clustering based on feature selection and semi-non-negative anchor graph factorization.

Multi-view clustering has garnered significant attention due to its capacity to utilize information from multiple perspectives. The concept of anchor graph-based techniques was introduced to manage large-scale data better. However, current methods rely on K-means or uniform sampling to select anchors in the original space. This results in a disjointed approach separating anchor selection and subsequent graph construction. Moreover, these methods typically require additional K-means or spectral clustering to derive labels, often leading to suboptimal outcomes. To address these challenges, we present a novel approach called Multi-view Clustering based on Feature Selection and Semi-Non-Negative Anchor Graph Factorization (MCFSAF). This method unifies feature selection, anchor and anchor graph learning, and semi-non-negative factorization of the anchor graph into a cohesive framework. Within this framework, the anchors and anchor graph are learned in the embedding space following feature selection, and the clustering indicator matrix is obtained via semi-non-negative factorization of the anchor graph in each view. By applying the minimization of the tensor Schatten p-norm, we can uncover complementary information across multiple views efficiently. This synergetic process of anchor selection, anchor graph learning, and indicator matrix updating can effectively enhance the clustering quality. Critically, the fused indicator matrix enables us to directly acquire clustering labels without requiring additional K-means, thereby significantly improving the stability of the clustering process. Our method is optimized via an alternating iterations algorithm. Comprehensive experimental evaluations underscore the superior performance of our approach.

Improved Aircraft Environmental Impact Segmentation via Metric Learning

Accurate modeling of aircraft environmental impact is pivotal to the design of operational procedures and policies to mitigate negative aviation environmental impact. Aircraft environmental impact segmentation is a process that clusters aircraft types that have similar environmental impact characteristics based on a set of aircraft features. This practice helps model a large population of aircraft types with insufficient aircraft noise and performance models and contributes to better understanding of aviation environmental impact. Through measuring the similarity between aircraft types, a distance metric forms the kernel of aircraft segmentation. Traditional ways of aircraft segmentation use plain distance metrics and assign equal weight to all features in an unsupervised clustering process. In this work, we utilize weakly supervised metric learning and partial information on aircraft fuel burn, emissions, and noise to learn weighted distance metrics for aircraft environmental impact segmentation. We show in a comprehensive case study that the tailored distance metrics can indeed make aircraft segmentation better reflect the actual environmental impact of aircraft. The metric learning approach can help refine a number of similar data-driven analytical studies in aviation.

Penerapan Data Mining Clustering terhadap Perekonomian di Kelurahan Sawah Lebar Menggunakan Algoritma K-Means

This study aims to analyze the economic conditions of the community in Sawah Lebar Village, Ratu Agung District, Bengkulu City, Bengkulu Province, using the data mining clustering method based on the K-Means algorithm. The K-Means algorithm is applied to group economic data of the community based on several variables, such as Employment Status, Home Ownership, Dependents, Monthly Income, Monthly Expenditure and Age of Head of Family. Through this clustering process, several groups (clusters) are produced that identify different economic patterns in the region, namely Cluster_0 (C0), Cluster_1 (C1) and Cluster_2 (C2). In order for the results of this study to be accurate, the data processing in this study uses the Rapidminer Studio Application. From the results of the data processing that has been carried out, it was found that for C0 there are 17 data, C1 there are 16 Data and C2 there are 17 Data. Then, for the recommendation of the community that is recommended to receive government assistance, it is the community that is in the lowest economic class, namely Cluster_1 (C1), Cluster_2 (C2) is the community with the highest economy, while the community in Cluster_0 (C0) is the middle economy.

Accelerating BRICS Economic Growth: AI-Driven Data Analytics for Informed Policy and Decision Making

This paper analyzes how the Artificial Intelligence (AI) and Machine Learning (ML) are bridging the gap between economic growth in the BRICS countries. BRICS countries are emerging economies that are challenged by increasing income inequality, industrial transformation and the need for infrastructure development. Driven by AI, this study applies data analytics to macroeconomic datasets, tracking down patterns and functional takeaways regarding policy formulation and strategic decision making. The research employs techniques, including predictive modeling, clustering, and natural language processing (NLP), in areas such as trade optimization, resource allocation and labour market analysis. Case examples document successful introduction of AI systems to solve critical economic problems, from increasing healthcare access to raising productivity in agriculture. The findings illustrate the role of AI and ML in helping BRICS policymakers to an informed, data driven development. The research puts AI as core to the process of economic advancement, a solution to developmental gaps and a driver for growth. This research contributes both to its practical outcomes and by providing insights into how AI and ML can solve the complex economic problems of emerging markets. The paper introduces predictive modeling, which anticipates economic trends based on past data and clustering which groups similar economic behaviors to find patterns as tools that are important in economic analysis. Further, Natural Language Processing (NLP) is covered as a highly effective approach to understand policy documents, news, and unstructured data to improve the ability to make decisions. By helping students, researchers, and policymakers understand these AI powered techniques that optimize trade, resource management and labor, these scalable solutions to sustainable development are available. This study touts data driven innovation as a critical means to solve global challenges, well-equipped readers with the skills and knowledge to leverage AI for economic progress in a geography of the dynamic and connected.

Analysis of the Evaluation Data of College Classroom Teaching Quality based on Fuzzy Mathematics Model

This paper mainly studies the problem of teaching quality evaluation. According to the data in the annex, the transformation of quantitative indicators and the evaluation of teaching quality are realized by establishing fuzzy mathematical model, grey correlation degree and other methods. For problem 1: according to the quantitative values of the eight evaluation indexes, a fuzzy mathematical model is established, and the comprehensive evaluation is transformed into a quantitative index to evaluate the teaching level of teachers. See the appendix for the sequence of serial numbers 4450-4469 in Annex 1. For question 2: considering the large amount of data in Annex 1, the courses that have participated in the evaluation more than 50 times are selected as the research object. Using hierarchical clustering processing, it is concluded that there are some differences in the teaching quality of each course in the school. According to the overall teaching quality, the seven teaching teams are: course 0097, course 0057, course 0322, course 0320, course 0486, course 0058, course 0138. For question 3: the grey correlation model is established according to the factors that affect the professionalism and objectivity of evaluation experts, such as teachers' professional titles, the number of evaluation sessions and the average number of sessions. The evaluation conclusions of the evaluation experts are given in the appendix. For question 4: the data in Annex 1 are sorted out in chronological order to obtain 46 evaluation objects and three evaluation indexes, namely, the number of evaluation sections, the average number of sections and the quantification of professional titles, for analysis. Then through the fitting of data processing, it is concluded that the teaching quality of the school has declined with the growth of time in the past three years. In response to question 5: according to the data in the annex, this paper puts forward the question "what is the relationship between the scores of teachers' comprehensive evaluation of teaching and students' comprehensive evaluation of learning in the evaluation of classroom teaching quality in Colleges and universities?" by fitting the fitting image with MATLAB, it is concluded that the scores of teachers' comprehensive evaluation of teaching and students' comprehensive evaluation of learning in the evaluation of classroom teaching quality in Colleges and universities are very similar.

Training Model of Students’ English Creative Thinking Ability Based on Big Data Analysis

In order to optimize the evaluation of students’ English creative thinking ability under the multimedia assisted teaching mode, a training model of students’ English Creative Thinking Ability Based on big data analysis is proposed. Establish a big data block fusion structure analysis model, analyze the fuzzy parameters of constraint indicators, obtain a quantitative table for the state allocation of evaluation constraint indicators, formulate an evaluation quantitative analysis scoring table, and use hierarchical structural reconstruction to construct the clustering model parameters for the evaluation of students’ English creative thinking ability training. Through the quantitative index feature analysis, the optimization design of teaching activities and teaching methods for the evaluation of the cultivation process of students’ English creative thinking ability are realized. In the multi-dimensional hierarchical structure parameter model, the parameter configuration of students’ English creative thinking ability training indicators is realized, and the evaluation characteristic index fusion clustering processing is carried out on the parameter configuration results to form a pair of classification prediction and index analysis model. According to the hierarchical distribution density and grid clustering of indicators, the evaluation of students’ English creative thinking ability under the multimedia-assisted teaching mode is realized. The results of empirical analysis show that the fuzzy parameter analysis ability of the constraint index of process evaluation using this method is strong, the evaluation results are accurate and reliable, and the reliability and confidence level of evaluation are improved.

A Network-Based Clustering Method to Ensure Homogeneity in Regional Frequency Analysis of Extreme Rainfall

The social impacts of extreme rainfall events are expected to intensify with climate change, making reliable statistical analyses essential. High quantile estimation requires substantial data; however, available records are sometimes limited. Additionally, finite data and variability across statistical models introduce uncertainties in the final estimates. This study addresses the uncertainty that arises when selecting parameters in Regional Frequency Analysis (RFA) by proposing a method to objectively identify statistically homogeneous regions. Station coordinates, elevation, annual mean rainfall, maximum annual rainfall, and l-skewness from 55 meteorological stations are selected to study annual maximum daily rainfall. These covariates are employed to investigate the interdependency of the covariates in Principal Component Analysis (PCA) as a preprocessing step in cluster analysis. Network theory, implemented through an iterative clustering process, is used in network creation where stations are linked based on the frequency of their co-occurrence in clusters. Communities are formed by maximizing the modularity index after creating a network of stations. RFA is performed in the final communities using L-moment theory to estimate regional and InSite quantiles. Quantile uncertainty is calculated through parametric bootstrapping. The application of PCA has a negligible effect on network creation in the study area. The results show that the iterative clustering approach with network theory ensures statistically created homogeneous regions, as demonstrated in Thessaly’s complex terrain for regionalisation of extreme rainfall.

Кластеризация вихрей Абрикосова в интертипных сверхпроводниках в присутствии центров пиннинга

In this work, the processes of clusterization and melting of a vortex lattice with an intertype potential of intervortex interaction were studied. Numerical modeling was performed within the framework of the Lawrence-Doniak model using the Monte Carlo method. It was shown that the structure obtained in the intertype superconductor is more stable to field and temperature changes than the vortex structure of a type II superconductor. The process of melting of the cluster lattice is shown, including in the presence of pinning centers. Differences in the processes of melting of the vortex lattice of a defective and defect-free superconductor are shown.

Advanced Hybrid Brain Tumor Segmentation in MRI: Elephant Herding Optimization Combined with Entropy-Guided Fuzzy Clustering

Accurate and early detection of brain tumors is essential for improving clinical outcomes and guiding effective treatment planning. Traditional segmentation techniques in MRI often struggle with challenges such as noise, intensity variations, and complex tumor morphologies, which can hinder their effectiveness in critical healthcare scenarios. This study proposes an innovative hybrid methodology that integrates advanced metaheuristic optimization and entropy-based fuzzy clustering to enhance segmentation precision in brain tumor detection. This method combines the nature-inspired Elephant Herding Optimization (EHO) algorithm with Entropy-Driven Fuzzy C-Means (EnFCM) clustering, offering significant improvements over conventional methods. EHO is utilized to optimize the clustering process, enhancing the algorithm’s ability to delineate tumor boundaries, while entropy-based fuzzy clustering accounts for intensity inhomogeneity and diverse tumor characteristics, promoting more consistent and reliable segmentation results. This approach was evaluated using the BraTS challenge dataset, a benchmark in the field of brain tumor segmentation. The results demonstrate marked improvements across several performance metrics, including Dice similarity, mean squared error (MSE), peak signal-to-noise ratio (PSNR), and the Tanimoto coefficient (TC), underscoring this method’s robustness and segmentation accuracy. By managing image noise and reducing computational demands, the EHO-EnFCM approach not only captures intricate tumor structures but also facilitates efficient image processing, making it suitable for real-time clinical applications. Overall, the findings reveal the potential of this hybrid approach to advance MRI-based tumor detection, offering a promising tool that enhances both accuracy and computational efficiency for medical imaging and diagnosis.

Adaptive Compression and Reconstruction for Multidimensional Medical Image Data: A Hybrid Algorithm for Enhanced Image Quality.

Spatial regions within images typically hold greater priority over adjacent areas, especially in the context of medical images (MI) where minute details can have significant clinical implications. This research addresses the challenge of compressing medical image dimensions without compromising critical information by proposing an adaptive compression algorithm. The algorithm integrates a modified image enhancement module, clustering-based segmentation, and a variety of lossless and lossy compression techniques. Edge enhancement contrast limited adaptive histogram equalization (EE-CLAHE) and 2D adaptive anisotropic diffusion filter are employed to enhance and denoise the images, followed by adaptive expectation maximization clustering (AEMC) for segmentation into regions of interest (ROI) and non-ROI. The clustering process is optimized utilizing fuzzy c-means (FCM) and Otsu thresholding. Subsequently, distinct compression schemes are applied to ROI and non-ROI regions, such as Coiflet + Haar, Coiflet + Daubecheis, modified SPIHT Huffman, EZW, and SPIHT algorithms, to ensure effective storage and transmission while preserving diagnostic details. Experimental results demonstrate that the combination of the modified SPIHT Huffman algorithm for ROI and EZW for non-ROI yields superior reconstruction quality across various measures, enabling comprehensive analysis of multi-dimensional images from MRI, CT, and X-ray modalities.

Classifying continuous GNSS stations using integrated machine learning

The development of Global Navigation Satellite Systems (GNSS) results in large spatial geodetic networks with a distinct range of accuracy. Thus, classification of the GNSS stations is needed to determine which stations are appropriate for geodetic applications. Additionally, advanced Machine Learning (ML) techniques have been proposed. However, ML algorithms may sometimes be less sensitive due to a lack of samples or anomalies in input data. Therefore, this study introduces an approach in which human-based supervision is integrated into ML processes to improve the ML model’s performance in classifying the continuous GNSS stations. The human factor influences the ML processes through two sampling strategies: “suggest-decide” and “correct-retrain”, where the accuracy of ML models will be improved via human-based corrections. The idea is that the unsupervised ML-based clustering techniques are driven by human-based supervision to create samples for training the supervised ML-based classification models. In this study, we develop a MATLAB app to automate the clustering and labeling processes. Our finding demonstrates that applying these sampling strategies can enhance the accuracy of the ML-based classification models from under 50 % up to ∼99 % after re-training. Also, this study categorizes almost 9000 continuous monitoring stations in the Nevada database, of which 1900 stations in Europe serve as samples for training the ML-based classification models. Furthermore, the methodologies developed in this study can be applied to warning systems, which utilize internal and external human resources to correct errors, address unusual situations, and provide timely feedback for better performance of ML-based forecasts.

Spatio-temporal hierarchical clustering of interval time series with application to suicide rates in Europe

In this paper, we investigate similarities of suicide rates in Europe, which are available as interval time series. For this aim, a novel spatio-temporal hierarchical clustering algorithm for interval time-series data is proposed. The spatial dimension is included in the clustering process to account for possible relevant information such as weather conditions, sunlight hours and socio-cultural factors. Our results indicate the presence of six main clusters in Europe, which almost overlap with the sunlight hours distribution. Differences between male and female suicide rates are also investigated.