Combining K-means Clustering Research Articles

Toward an Efficient Hybrid Clustering and Classification Approach for Fake News Detection in Social Network

Objectives: The spread of fake news on social media has become a pressing issue in recent times. Despite various organizations' efforts to address this problem, it continues to persist, necessitating finding more effective solutions. This study implements a machine learning-based approach for identifying fake news on social media with improved accuracy. Methods : The study's methodology utilizes a combination of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to extract semantic features from news articles. Then, a hybrid clustering and classification approach is used, which combines K-means clustering and Artificial Neural Network (ANN) classifiers. Data used is a secondary data consisting of 23481 world news articles obtained from the Reuters.com website and 21417 unreliable news from polifact.com. During the training process, the number of units and layers on the model was tuned to optimize model performance. The model was compared to other baseline models such as KNN, SVM, Decision tree, and Boosted decision tree to establish the best performing model. Findings: The results from the two algorithms were weighted to final classifications using Bayesian probability theory. The proposed approach achieved an accuracy of 99.78%, a sensitivity of 100%, and specificity equal to 99.73%. The model's precision is 99.74%, indicating its ability to identify fake news. The F-score of the approach is 99.87%, indicating that the model strikes a good balance between correctly classifying fake news articles and reliable news articles. The approach outperformed other machine learning classifiers, including KNN, SVM, Decision Tree, and Boosted Decision Tree. Novelty : The study applies a hybrid approach with a classification and clustering algorithm to improve detection of fake news on social media, the approach is tested with varied real-world datasets to establish its robustness under different vocabularies and vocabulary sizes. Keywords: Artificial Intelligence, Machine Learning, Deep Learning, Hybrid clustering approach, Classification, CNN, LSTM, Boosted Decision Tree, Social platforms, K-means

Dominant Color Detection For Online Fashion Retrievals

This paper introduces a novel approach aimed at efficiently extracting dominant colors from online fashion images. The method addresses challenges related to detecting overlapping objects and computationally expensive methods by combining K-means clustering and graph-cut techniques into a framework. This framework incorporates an adaptive weighting strategy to enhance color extraction accuracy. Additionally, it introduces a two-phase fashion apparel detection method called YOLOv4, which utilizes U-Net architecture for clothing segmentation to precisely separate clothing items from the background or other elements. Experimental results show that K-means with YOLOv4 outperforms K-means with the U-Net model. These findings suggest that the U-Net architecture and YOLOv4 models can be effective methods for complex image segmentation tasks in online fashion retrieval and image processing, particularly in the rapidly evolving e-commerce environment.

GreenRoute Ballari: A Hybrid Approach for Optimizing Waste Collection Routes Using K-Means Clustering and Genetic Algorithm

Abstract: The exponential growth of the population in Ballari has exacerbated the challenge of municipal waste management, necessitating the development of efficient waste collection strategies to address environmental and social concerns. In response to that our system introduces a hybrid approach that employs K-means clustering and Genetic Algorithm, for optimizing waste collection routes tailored to the unique waste generation profiles of the region. By utilizing Geographic Information System (GIS) data that is in the form of GeoJSON files containing the spatial distribution of waste collection points, we propose a methodology that combines K-means clustering and Genetic Algorithm to develop intelligent and fuel-efficient routes. Our system not only benefits municipal authorities in resource deployment and planning but also aids waste carrier vehicle drivers in prioritizing routes. Our findings contribute to the transformation of Ballari City towards eco-friendly and smart urban development

Automatic simulation of electrochemical sensors by machine learning for drugs quantification

Multiple drug concentrations concurrent detection and quantification based on electrochemical sensors are of great importance for therapeutic drug monitoring (TDM) and the development of personalized therapy. Cyclic voltammogram (CV) results obtained by electrochemical sensors can be used to offer quantitative information about drug concentrations. Several approaches have been proposed for single-concentration quantification based on CV results and machine learning (ML) models with lower training difficulty. However, insufficient measured dataset hinders the application of diverse large-scale ML algorithms in this field. A new method for automatic parameter estimation by ML of measured CV samples is here illustrated with the aim to generate a large simulated dataset for generalized drug concentration quantification model training in this paper. We present an ML-based approach that combines k-means clustering, polynomial regression, and Gaussian Mixture Model (GMM), which automates parameter estimation and simulation using peak detection, baseline subtraction, and peak Gaussian fitting with a small number of measurements. Large simulation datasets constructed on the basis of the estimation results open the possibility of training ML models for more generalized drug concentration quantification. The simulated dataset is processed to assess the efficiency of the proposed method. The Mean Average Percentage Error (MAPE) was 0.32% for etoposide (ETO) and 4.78% for methotrexate (MTX).

Breast cancer classification based on hybrid machine learning model

This study proposes a hybrid model that combines K-Means clustering and Random Forest classification as an approach for breast cancer classification. The objective is to exploit the advantages of unsupervised clustering and supervised classification techniques to enhance the accuracy and robustness of classification models. The dataset underwent preprocessing procedures encompassing the handling of missing values, feature normalization, and feature selection. Missing values were addressed through appropriate methods, and features were scaled and selected based on variance threshold or correlation analysis. Subsequently, K-Means clustering was applied to the preprocessed data to assign cluster labels to each sample. The study then proceeded to train a Random Forest classifier by incorporating both the cluster labels and the raw gene eigenvalues as mixed features. This integration of gene expression values and cluster labels provides supplementary information to the classifier, enabling the capture of more intricate patterns within the data. The Random Forest classifier was trained using optimized parameters determined through parameter tuning, including the number of trees, maximum depth, and minimum number of split samples. Extensive experiments and evaluations conducted in this study revealed that the hybrid model outperformed the standalone Random Forest classification. The incorporation of K-Means clustering facilitated the discovery of underlying data structures and patterns, ultimately enhancing the classifier's discriminatory ability. The hybrid model exhibited superior accuracy, precision, recall, and F1 scores, demonstrating its efficacy in accurately classifying breast cancer samples.

Kmeans-SMOTE Integration for Handling Imbalance Data in Classifying Financial Distress Companies using SVM and Naïve Bayes

Imbalanced data presents significant challenges in machine learning, leading to biased classification outcomes that favor the majority class. This issue is especially pronounced in the classification of financial distress, where data imbalance is common due to the scarcity of such instances in real-world datasets. This study aims to mitigate data imbalance in financial distress companies using the Kmeans-SMOTE method by combining Kmeans clustering and the synthetic minority oversampling technique (SMOTE). Various classification approaches, including Nave Bayes and support vector machine (SVM), are implemented on a Kaggle financial distress data set to evaluate the effectiveness of Kmeans-SMOTE. Experimental results show that SVM outperforms Nave Bayes with impressive accuracy (99.1%), f1-score (99.1%), area under precision recall (AUPRC) (99.1%), and geometric mean (Gmean) (98.1%). On the basis of these results, Kmeans-SMOTE can balance the data effectively, leading to a quite significant improvement in performance.

Classification of building complex for the large-scale construction of distributed photovoltaics in urban buildings

The large-scale development of urban building photovoltaics (PV) has become an important avenue for cities to achieve building energy conservation, emission reduction, and carbon neutrality. Assessing the solar energy potential of urban buildings plays a crucial role in the installation and overall planning of urban building PV systems. However, due to the vast number and diverse types of urban buildings, individual building calculations can be both labor-intensive and inefficient. Therefore, batch processing of buildings after classification is a feasible approach. Previous assessments of solar energy potential in building clusters mainly focused on typical clusters or standard geometric areas, but all urban building types were covered. Based on this, this paper proposes a method for defining Building Complexes with a Single Land Use Nature (BCSLUN) by combining geographical distribution and urban land use characteristics. Furthermore, population density, building density, compactness, and land use characteristics are selected as features, and a method that combines K-means clustering and random forest classification is proposed to complete the classification of building complexes. Taking the central urban area of Xi'an, China as an example, 5,804 building complexes are classified into four building types. The cross-validation classification accuracy is 0.99, and the Kappa coefficient is 0.94. Based on Rhino modeling, the global solar radiation on the surface of each building complex was calculated. The annual surface global radiation of the four building types is 11,442.73 MWh, 2,734.71 MWh, 13,923.81 MWh, and 67,215 MWh. The violin-box plot shows a significant classification effect, which proves the rationality of the proposed classification method. Finally, a potential map based on solar radiation from urban building surfaces is generated. Building complex classification can quickly and accurately estimate the solar energy potential of urban buildings, providing decision-making support for the large-scale construction of urban building PV. Moreover, this method can be easily applied to other cities.

Automatic high depth-resolvability inspection of TBC debonding defects using nonlinear frequency modulation photothermal radar

In this article, a novel nonlinear frequency modulation (NLFM) photothermal radar whose instantaneous frequency curve is a concave quadratic function is proposed and further combines K-means clustering image segmentation algorithm to achieve automatic high depth-resolvability inspection of debonding defects in thermal barrier coating (TBC) structure. Firstly, a 3-dimensional (3D) photothermal finite element (FE) model of TBC structure is established, where the extrapolated Beer-Lambert model is introduced to describe the spatially volumetric heat source within the ceramic coating. Afterwards, the superior depth resolvability of the proposed NLFM photothermal radar is demonstrated compared to conventional waveforms using the established FE model. Meanwhile, its depth resolvability dependence on the thickness and thermal diffusivity of the ceramic coating is also investigated. Finally, the photothermal experiments of the TBC structure containing artificial debonding defects are conducted, and the influence of the coating thickness and metal substrate on the non-destructive testing (NDT) detection of TBC structure by NLFM photothermal radar combining K-means clustering is also experimentally studied. The results reveal that NLFM photothermal radar combining K-means clustering can achieve automatic high depth-resolvability inspection of debonding defects in TBC structure and the coating thickness and metal substrate have also a great influence on the photothermal NDT of the debonding defects of TBC structure.

STGNNks: Identifying cell types in spatial transcriptomics data based on graph neural network, denoising auto-encoder, and [formula omitted]-sums clustering

Background:Spatial transcriptomics technologies fully utilize spatial location information, tissue morphological features, and transcriptional profiles. Integrating these data can greatly advance our understanding about cell biology in the morphological background. Methods:We developed an innovative spatial clustering method called STGNNks by combining graph neural network, denoising auto-encoder, and k-sums clustering. First, spatial resolved transcriptomics data are preprocessed and a hybrid adjacency matrix is constructed. Next, gene expressions and spatial context are integrated to learn spots’ embedding features by a deep graph infomax-based graph convolutional network. Third, the learned features are mapped to a low-dimensional space through a zero-inflated negative binomial (ZINB)-based denoising auto-encoder. Fourth, a k-sums clustering algorithm is developed to identify spatial domains by combining k-means clustering and the ratio-cut clustering algorithms. Finally, it implements spatial trajectory inference, spatially variable gene identification, and differentially expressed gene detection based on the pseudo-space-time method on six 10x Genomics Visium datasets. Results:We compared our proposed STGNNks method with five other spatial clustering methods, CCST, Seurat, stLearn, Scanpy and SEDR. For the first time, four internal indicators in the area of machine learning, that is, silhouette coefficient, the Davies-Bouldin index, the Caliniski-Harabasz index, and the S_Dbw index, were used to measure the clustering performance of STGNNks with CCST, Seurat, stLearn, Scanpy and SEDR on five spatial transcriptomics datasets without labels (i.e., Adult Mouse Brain (FFPE), Adult Mouse Kidney (FFPE), Human Breast Cancer (Block A Section 2), Human Breast Cancer (FFPE), and Human Lymph Node). And two external indicators including adjusted Rand index (ARI) and normalized mutual information (NMI) were applied to evaluate the performance of the above six methods on Human Breast Cancer (Block A Section 1) with real labels. The comparison experiments elucidated that STGNNks obtained the smallest Davies-Bouldin and S_Dbw values and the largest Silhouette Coefficient, Caliniski-Harabasz, ARI and NMI, significantly outperforming the above five spatial transcriptomics analysis algorithms. Furthermore, we detected the top six spatially variable genes and the top five differentially expressed genes in each cluster on the above five unlabeled datasets. And the pseudo-space-time tree plot with hierarchical layout demonstrated a flow of Human Breast Cancer (Block A Section 1) progress in three clades branching from three invasive ductal carcinoma regions to multiple ductal carcinoma in situ sub-clusters. Conclusion:We anticipate that STGNNks can efficiently improve spatial transcriptomics data analysis and further boost the diagnosis and therapy of related diseases. The codes are publicly available at https://github.com/plhhnu/STGNNks.

Application of Unsupervised/Supervised Machine Learning to Harmful Gas Detection/Classification Based on Mutually-Interacting Semiconductor Metal Oxide Gas Sensors

Chemical and/or physical sensors are employed at domestic and industrial sites with the aim to perform a variety of functions, such as prediction, forecasting, prognostics, remaining useful life estimation, anomaly detection, and trend analysis. The sensing information contain the analog signals originating from gas-generated signals, pressure, shear, strain, torsion, temperature, humidity, illumination, electromagnetic radiation, sound, and vibration. Chemical gas sensors can be categorized mainly as semiconductor metal oxide (SMO), electrochemical, or photo-ionization detection (PID) sensors. SMO gas sensors are based on adsorption and desorption of gas molecules onto/from semiconductor metal oxides and however, suffer from the low selectivity to diverse gas species, i.e., the mutually interacting features.Here, we report a high-performance machine learning strategy for gas detection/discrimination against harmful gas combination and adopt the synergistic integration of supervised and unsupervised learning by exploiting a SMO gas sensor array. A 4 SMO sensor array was constructed for detecting carbon monoxide and ethyl alcohol (C2H5OH) mixtures using 15 different gas combinations. Gas detection/discrimination performance was probed in terms of different numbers of gas sensor integrated as an array and machine learning algorithms. Unsupervised K-Means clustering was successfully applied to the rational identification of the similarity features of targeted gases among 4 different groups, which are listed as matrix gas, two single-component gases, and one two-gas by employing only unlabeled voltage-based gas sensing information. Detailed classification was performed through a multitude of supervised algorithms, i.e., 2-layer artificial neural networks (ANNs), 4-layer deep neural networks (DNNs), 1-dimensional convolutional neural networks (1D CNNs), and 2-dimensional convolutional neural networks (2D CNNs). The numerical-based DNNs and image-based CNNs are evaluated to be excellent approaches for gas detection/classification, as corroborated by the highest accuracy and lowest loss parameters.Through the systematic investigation on the influence of the number of sensors on the arrayed gas sensor system, the applicability of machine learning methodology to an arrayed SMO gas sensor system is justified by the four unique features, i.e., i) a data augmentation methodology, ii) machine learning approach of combining K-means clustering and neural networks, and iii) a systematic approach to optimized sensor combinations, potentially recommending the minimized sensor systems based on chemical gas sensors against harmful gas species. Even two SMO sensor combinations are shown to be highly effective in performing gas discrimination against harmful gas species assisted through either numeric-based DNNs or image-based 1D CNNs, overcoming the simple clustering proposed through the unsupervised K-means clustering.

Instant testing and non-contact diagnosis for photovoltaic cells using K-means clustering and associated hyperspectral imaging

Renewable energy, particularly solar energy, has experienced remarkable growth in recent years. However, the integrity of solar photovoltaic (PV) cells can degrade over time, necessitating non-destructive testing and evaluation (NDT-NDE) for quality control during production and in-service inspection. Hyperspectral (HS) imaging has emerged as a promising technique for defect identification in PV cells based on their spectral signatures. This study utilizes a HS imager to establish a diffuse reflectance spectra signature for two groups of PV cells: working and non-working. A non-contact photoluminescence imaging-based methodology is employed, using a halogen lamp as an illumination source to replicate sunlight. Our findings reveal that non-working PV regions can be differentiated from working regions within the 400–600 nm wavelength range, with an optimal candidate peak frequency of 450 nm. To accurately group active PV regions in the constructed HS images at 450 nm, we employ an image processing strategy that combines K-means clustering (K-mc) with contour delineation. Specifically, K-mc with K = 8 is used to efficiently and precisely group active PV regions. We demonstrate the effectiveness of this proposed approach and compare it with traditional infrared (IR) imaging techniques. This imaging clustering approach can be implemented using a conventional camera and a 450 nm wavelength filter for NDT-NDE on exterior-mounted PV panels. Overall, the proposed HS imaging technique, coupled with K-mc, offers a rapid and effective means of identifying defects in PV cells, outperforming conventional IR imaging techniques. This advancement contributes to increased efficiency and extended lifespan of solar PV panels.

Public Management Model Based on Mathematical Equation of Statistics of Probability Theory

Abstract Firstly, this paper establishes a cloud computing public opinion management business model based on distributed collection analysis, consultative decision and concurrent collaborative behavior. At the same time, this paper studies the public opinion management system from the logical framework, network structure and security index. Then this paper uses the weak correlation mining technology of probability theory statistics to establish a public health and safety public opinion prediction system composed of three dimensions. The content includes three levels of public opinion bonding, thermal and trend. Then, this paper forecasts the development of public opinion events by combining K-Means clustering and grey correlation degree. The results show that this method is suitable for public opinion prediction in public health and safety.

CUDA-bigPSF: An optimized version of bigPSF accelerated with graphics processing Unit

Accurate and fast short-term load forecasting is crucial in efficiently managing energy production and distribution. As such, many different algorithms have been proposed to address this topic, including hybrid models that combine clustering with other forecasting techniques. One of these algorithms is bigPSF, an algorithm that combines K-means clustering and a similarity search optimized for its use in distributed environments. The work presented in this paper aims to improve the time required to execute the algorithm with two main contributions. First, some of the issues of the original proposal that limited the number of cores simultaneously used are studied and highlighted. Second, a version of the algorithm optimized for Graphics Processing Unit (GPU) is proposed, solving the previously mentioned issues while taking into account the GPU architecture and memory structure. Experimentation was done with seven years of real-world electric demand data from Uruguay. Results show that the proposed algorithm executed consistently faster than the original version, achieving speedups up to 500 times faster during the training phase.

Development of Forest Carbon Sequestration Model

Firstly, the factors affecting carbon sequestration are analyzed based on tree age, precipitation, temperature and light. On this basis, the carbon sequestration model is developed by combining K-means clustering, regression analysis verification and grey prediction model. Finally, the carbon sequestration model is used to predict the trend of carbon sequestration content of six vegetation types in the next two years. The research results provide guidance for sequestration of carbon dioxide.

Machine Learning-Assisted Gas-Specific Fingerprint Detection/Classification Strategy Based on Mutually Interactive Features of Semiconductor Gas Sensor Arrays

A high-performance machine learning-assisted gas sensor strategy based on the integration of supervised and unsupervised learning with a gas-sensitive semiconductor metal oxide (SMO) gas sensor array is introduced. A 4-SMO sensor array was chosen as a test sensor system for detecting carbon monoxide (CO) and ethyl alcohol (C2H5OH) mixtures using 15 different combinations. Gas sensing detection/classification was performed with different numbers of gas sensor and machine learning algorithms. K-Means clustering was successfully employed to rationally identify the similarity features of targeted gases among 4 different groups, i.e., matrix gas, two single-component gases, and one two-gas mixture, based on only unlabeled voltage-based gas sensing information. Detailed classification was performed through a multitude of supervised algorithms, i.e., 2-layer artificial neural networks (ANNs), 4-layer deep neural networks (DNNs), 1-dimensional convolutional neural networks (1D CNNs), and 2-dimensional CNNs (2D CNNs). The numerical-based DNNs and image-based CNNs are shown to be excellent approaches for gas detection and classification, as indicated by the highest accuracy and lowest loss indicators. Through the analysis of the influence of the number of sensors on the arrayed gas sensor system, the application of machine learning methodology to an arrayed gas sensor system demonstrates four unique features, i.e., a data augmentation methodology, machine learning approach of combining K-means clustering and neural networks, and a systematic approach to optimized sensor combinations, potentially leading to the practical sensor networks based on chemical sensors. Even two SMO sensor combinations are shown to be highly effective in gas discrimination against diverse gas environments assisted through numeric-based DNNs and image-based 1D CNNs, overcoming the simple clustering proposed through the unsupervised K-means clustering.

A Sparsity Adaptive Algorithm to Recover NB-IoT Signal From Legacy LTE Interference

As a forerunner in 5G technologies, Narrowband Internet of Things (NB-IoT) will be inevitably coexisting with the legacy Long-Term Evolution (LTE) system. Thus, it is imperative for NB-IoT to mitigate LTE interference. By virtue of the strong temporal correlation of the NB-IoT signal, this letter develops a sparsity adaptive algorithm to recover the NB-IoT signal from legacy LTE interference, by combining K-means clustering and sparsity adaptive matching pursuit (SAMP). In particular, the support of the NB-IoT signal is first estimated coarsely by K-means clustering and SAMP algorithm without sparsity limitation. Then, the estimated support is refined by a repeat mechanism. Simulation results demonstrate the effectiveness of the developed algorithm in terms of recovery probability and bit error rate, compared with competing algorithms.

Image segmentation using K-means clustering, Gabor filter and moving mesh method

ABSTRACT Some medical images such as blood vessels or nerve canal are very narrow, it is hardly segmented. In this paper, we propose a two-dimensional image segmentation algorithm combining K-means clustering, Gabor filter and moving mesh method. The mesh is moved adaptively concentrated in steep region and the location feature is added to our clustering algorithm. Moreover, the image feature values are obtained by the convolution operation between the image and Gabor filter. The advantage of this approach is that the segmentation result of the K-means clustering algorithm can be improved by constantly adding new features such as location, pixel or Gabor filter. The experiments indicate that our algorithms can quickly segment medical and crack images without compromising the quality of the results. The results demonstrated that the proposed method achieved precision 94.31%, recall 97.84%, F1 score 95.96%.

A failure boundary exploration and exploitation framework combining adaptive Kriging model and sample space partitioning strategy for efficient reliability analysis

Surrogate model-based methods have gradually become a vital method to assess reliability. However, the existing methods usually ignore the memory problems of matching candidate samples with the level of failure probability, which leads to inefficiency and even restricts their applicability. Therefore, this work combining the adaptive Kriging model and sample space partitioning strategy proposes a failure boundary exploration and exploitation framework (FBEEF), which divides the construction process of the adaptive Kriging model into two phases using different candidate samples to enrich training samples. In the exploration phase, a sample space partitioning strategy combining K-means clustering and slice sampling is employed to obtain several subsets and static candidate samples. In the exploitation phase, the approximate distances between the static candidate samples and the failure boundary are calculated to identify important subsets, whose samples are named dynamic candidate samples. Furthermore, a new stopping criterion is developed by combining leave-one-out method and weighted simulation method. To improve the efficiency of FBEEF Monte Carlo simulation or Importance Sampling is selected to estimate the final failure probability. Five examples were analyzed to test the effectiveness of FBEEF, and the results show that FBEEF can obtain good results with fewer training samples and lower analysis time.

Combining K-Means Clustering and Random Forest to Evaluate the Gas Content of Coalbed Bed Methane Reservoirs

The accurate calculation of the gas content of coalbed bed methane (CBM) reservoirs is of great significance. However, due to the weak correlation between the logging response of coalbed methane reservoirs and the gas content parameters and strong nonlinear characteristics, it is difficult for conventional gas content calculation algorithms to obtain more reliable results. This paper proposes a CBM reservoir gas content assessment method combining K-means clustering and random forest. The K-means clustering is used to divide the reservoirs and distinguish the types to establish a random forest model. Judging from the evaluation effect of the research block, the prediction accuracy of the new method is significantly higher than that of the original method, and more accurate gas content prediction values can be obtained for different types of reservoirs. Studies have shown that this method can help the gas content evaluation of CBM reservoirs, improve the accuracy of gas content evaluation, and better support the exploration and development of CBM reservoirs. The results of this study show that the random forest method based on clustering can effectively distinguish the relationship between different logging responses and gas content. On this basis, the random forest algorithm modeling can effectively characterize the complex relationship between gas content and logging curve response. In the case of poor correlation between gas content and logging curve, the gas content of the reservoir can also be accurately calculated.

A new model for learning-based forecasting procedure by combining k-means clustering and time series forecasting algorithms.

This paper aims to propose a new model for time series forecasting that combines forecasting with clustering algorithm. It introduces a new scheme to improve the forecasting results by grouping the time series data using k-means clustering algorithm. It utilizes the clustering result to get the forecasting data. There are usually some user-defined parameters affecting the forecasting results, therefore, a learning-based procedure is proposed to estimate the parameters that will be used for forecasting. This parameter value is computed in the algorithm simultaneously. The result of the experiment compared to other forecasting algorithms demonstrates good results for the proposed model. It has the smallest mean squared error of 13,007.91 and the average improvement rate of 19.83%.