Means Clustering Technique Research Articles

Wound Tissue Segmentation and Classification Using U-Net and Random Forest

Analysing wound tissue is a crucial research field for assessing the progression of wound healing. Wounds exhibit certain attributes concerning colour and texture, although these features can vary among different wound images. Research in this field serves multiple purposes, including confirming the presence of chronic wounds, identifying infected wounds, determining the origin of the wound and addressing other factors that classify and characterise various types of wounds. Wounds pose a substantial health concern. Currently, clinicians and nurses mainly evaluate the healing status of wounds based on visual examination. This paper presents an outline of digital image processing and traditional machine learning methods for the tissue analysis of chronic wound images. Here, we propose a novel wound tissue analysis system that consists of wound image pre-processing, wound area segmentation and wound analysis by tissue segmentation. The wound area is extracted using a simple U-Net segmentation model. Granulation, slough and necrotic tissues are the three primary forms of wound tissues. The [Formula: see text]-means clustering technique is employed to assign labels to tissues. Within the wound boundary, the tissue classification is performed by applying the Random Forest classification algorithm. Both segmentation (U-Net) and classification (Random Forest) models are trained, and the segmentation gives 99% accuracy, and the classification model gives 99.21% accuracy.

Privacy Preserving Using K Member Gaussian Kernel Fuzzy C Means and Self Adaptive Honey Badger for Online Social Networks

An online social network (OSN) gives users a strong platform to communicate and exchange information. Protecting publicly published data from individual identification is the primary problem in sharing social network databases. The most popular method for protecting privacy is anonymizing data, which involves deleting or altering some information while maintaining as much of the original data as feasible. This work presents a combination anonymizing algorithm, which is based on k member Gaussian kernel fuzzy c means clustering and self-adaptive honey badger optimization technique (KGKFCM-SAHBO). As part of the suggested anonymization process, the various users are divided into C clusters, each of which has at least K users, using a K-member Gaussian kernel fuzzy c means clustering technique. After that, the primary clusters are further optimized using the self-adaptive honey badger optimization approach (SAHBO) to further anonymize the data and network graph. Using the Yelp dataset, the experimental findings demonstrate the efficiency of the proposed model and evaluate a number of parameters, including execution time, degree of anonymization, and information loss. The experimental results show that the proposed strategy reduced information loss and improved the degree of anonymization when compared to existing methodologies.

Study of the Impact of Prosopis juliflora on Soil moisture and Humidity

This paper aims to present the impact of invasive alien species on soil moisture over the other native species. Prosopis juliflora is a majestic, non-native species that grows abundantly over the arid and semi-arid regions especially over the southern part of Indian sub-continent. These non-native, invasive species are modifying the native bio diversity, natural irrigation and local ecosystem by alternating the soil fertility measures such as soil moisture and relative humidity of the soil. In this study, the impact of the non-native species and native species on soil moisture and atmospheric humidity are considered for the analysis of the effect of the Prosopis juliflora and the variation in the conditions are visualized using the k- means clustering technique. The soil moisture at a depth of 2 meters and 1 meter and the air humidity is compared for regions in the presence of the nonnative species and in the absence of nonnative species. The results indicate that the non-native species has a negative impact on the cultivating land, spoiling its fertility measures by reducing its soil moisture content. The study provides evidence for the need to take necessary actions to eradicate species such as Prosopis juliflora and in turn to protect the native vegetation and develop management strategies to control the spread of the species.

Handling Massive Sparse Data in Recommendation Systems

Collaborative filtering-based recommendation systems have become significant in various domains due to their ability to provide personalised recommendations. In e-commerce, these systems analyse the browsing history and purchase patterns of users to recommend items. In the entertainment industry, collaborative filtering helps platforms like Netflix and Spotify recommend movies, shows and songs based on users’ past preferences and ratings. This technology also finds significance in online education, where it assists in suggesting relevant courses and learning materials based on a user’s interests and previous learning behaviour. Even though much research has been done in this domain, the problems of sparsity and scalability in collaborative filtering still exist. Data sparsity refers to too few preferences of users on items, and hence it would be difficult to understand users’ preferences. Recommendation systems must keep users engaged with fast responses, and hence there is a challenge in handling large data as these days it is growing quickly. Sparsity affects the recommendation accuracy, while scalability influences the complexity of processing the recommendations. The motivation behind the paper is to design an efficient algorithm to address the sparsity and scalability problems, which in turn provide a better user experience and increased user satisfaction. This paper proposes two separate, novel approaches that deal with both problems. In the first approach, an improved autoencoder is used to address sparsity, and later, its outcome is processed in a parallel and distributed manner using a MapReduce-based [Formula: see text]-means clustering algorithm with the Elbow method. Since the [Formula: see text]-means clustering technique uses a predetermined number of clusters, it may not improve accuracy. So, the elbow method identifies the optimal number of clusters for the [Formula: see text]-means algorithm. In the second approach, a MapReduce-based Gaussian Mixture Model (GMM) with Expectation-Maximization (EM) is proposed to handle large volumes of sparse data. Both the proposed algorithms are implemented using MovieLens 20M and Netflix movie recommendation datasets to generate movie recommendations and are compared with the other state-of-the-art approaches. For comparison, metrics like RMSE, MAE, precision, recall, and F-score are used. The outcomes demonstrate that the second proposed strategy outperformed state-of-the-art approaches.

Construction of agricultural product consumer group portrait and analysis of precision marketing strategies based on K-means cluster analysis

<p>In today’s technologically advanced landscape, where data flows in unparalleled volumes, the power of data mining stands out as a transformative force. Its capabilities extend beyond mere analysis; data mining can be harnessed to create intricate and detailed profiles of various consumer groups. This is particularly pertinent to the agricultural sector, which has long grappled with challenges of reaching its consumers effectively. The quest to refine and elevate marketing strategies for agricultural products amidst this data deluge led us to a methodical approach. We initiated this by sourcing and cleansing customer analysis datasets from Baidu AI, a leading platform in the realm of artificial intelligence and data analytics. Such a foundational step ensured that the data underpinning our analysis was robust and free of inconsistencies. The subsequent analytical journey, comprising rigorous data exploration and the utilization of the <em>K</em>-means clustering method, allowed us to dissect and segment the vast data pool. Through this, we crafted a comprehensive consumer profile that is tailor-made for agricultural product consumption. Such segmentation offers invaluable insights, paving the way for marketers and producers to understand their audience’s nuances. Our research findings highlight the remarkable prowess of the <em>K</em>-means clustering technique. When underpinned by sophisticated intelligent algorithms, it doesn’t just cluster data; it offers a pathway to identify distinct customer segments, shed light on core product offerings that resonate with them, and sculpt effective marketing strategies. By integrating these insights with modern media channels, we can craft marketing narratives that echo the desires and needs of the target demographic. Such a precision-driven approach ensures a symbiotic relationship where products align seamlessly with customer preferences. Beyond meeting immediate needs, this alignment has broader implications. It allows enterprises to tap into the vast opportunities presented by the digital era, positioning them on a trajectory of sustained growth and relevance in an ever-evolving market landscape.</p>

Hybridized Deep Neural Network Using Adaptive Rain Optimizer Algorithm for Multi-Grade Brain Tumor Classification of MRI Images

Classification of brain tumor is highly significant in the medical field in real-world to improve the progress of treatments. The seriousness behind the tumors are normally graded based on the size into grade I, grade II, grade III and grade IV. This is where the process of multi-grade brain tumor classification gains attention. Thus, the article focusses on classifying the brain MRI images into four different grades by proposing a novel and a very efficient classification strategy with high accuracy. The acquired images are pre-processed with the help of an Extended Adaptive Wiener Filter (EAWF) and then segmented using the piecewise Fuzzy C- means Clustering (piFCM) technique. Then the most ideal features such as the texture, intensity and shape features that can best explain the growth of tumors are extracted using the Local Binary Pattern (LBP) and the Hybrid Local Directional Pattern with Gabor Filter (HLDP-GF) techniques. After extracting the ideal features, the Manta Ray Foraging Optimization (MRFO) method has been introduced to optimally select the most relevant features. Finally, a Hybrid Deep Neural Network with Adaptive Rain Optimizer Algorithm (HDNN- AROA) is proposed to classify the grades of brain tumors with high accuracy and efficiency. The proposed technique has been compared with the existing state-of-the-art techniques relevant to brain tumor classification in terms of accuracy, precision, recall and dice similarity coefficient to prove the overall efficiency of the system.

Recommendation Learning System Model for Children with Autism

Autism spectrum disorder (ASD), is a neurological developmental disorder. It affects how people communicate and interact with others, as well as how they behave and learn. The symptoms and signs appear when a child is very young. Derived with increased usage of machine learning procedure in the medicinal analysis investigations. In this paper, our objective is to find out the most significant attributes and automate the process using classification techniques and pattern clustering using K-means clustering. We have analyzed ASD datasets of children towards determining the best performance of classifier for these binary datasets considering recall, precision, accuracy and classification errors. For this purpose only we use classifier along with a training model with the deep learning technique. In this deep neural network the different types of patterns have been obtained, and then the entire autism dataset has been trained using this classification technique. After that, the group of patterns has been formed using K- means clustering technique. Then these grouped patterns have been sent to the stochastic gradient descent (SGD) for obtaining the classification with enhanced accuracy, and after this, the regression process is done, then by this classified output, the recommendation learning system model is given for autism affected children. The need for recommendation for ASD is due to slow functioning of the brain and change in personal characteristic of the affected children, so it is necessary for recommendation learning model in ASD.

Two accelerated isogeometric boundary element method formulations: fast multipole method and hierarchical matrices method

This work presents two fast isogeometric formulations of the Boundary Element Method (BEM) applied to heat conduction problems, one accelerated by Fast Multipole Method (FMM) and other by Hierarchical Matrices. The Fast Multipole Method uses complex variables and expansion of fundamental solutions into Laurant series, while the Hierarchical Matrices are created by low rank CUR approximations from the k−Means clustering technique for geometric sampling. Both use Non-Uniform Rational B-Splines (NURBS) as shape functions. To reduce computational cost and facilitate implementation, NURBS are decomposed into Bézier curves, making the isogeometric formulation very similar to the conventional BEM. A description of the hierarchical structure of the data and the implemented algorithms are presented. Validation is performed by comparing the results of the proposed formulations with those of the conventional BEM formulation. The computational cost of both formulations is analyzed showing the advantages of the proposed formulations for large scale problems.

Detection of Emotional Sensitivity Using fNIRS Based Dynamic Functional Connectivity.

In this study, we proposed an analytical framework to identify dynamic task-based functional connectivity (FC) features as new biomarkers of emotional sensitivity in nursing students, by using a combination of unsupervised and supervised machine learning techniques. The dynamic FC was measured by functional Near-Infrared Spectroscopy (fNIRS), and computed using a sliding window correlation (SWC) analysis. A k -means clustering technique was applied to derive four recurring connectivity states. The states were characterized by both graph theory and semi-metric analysis. Occurrence probability and state transition were extracted as dynamic FC network features, and a Random Forest (RF) classifier was implemented to detect emotional sensitivity. The proposed method was trialled on 39 nursing students and 19 registered nurses during decision-making, where we assumed registered nurses have developed strategies to cope with emotional sensitivity. Emotional stimuli were selected from International Affective Digitized Sound System (IADS) database. Experiment results showed that registered nurses demonstrated single dominant connectivity state of task-relevance, while nursing students displayed in two states and had higher level of task-irrelevant state connectivity. The results also showed that students were more susceptive to emotional stimuli, and the derived dynamic FC features provided a stronger discriminating power than heart rate variability (accuracy of 81.65% vs 71.03%) as biomarkers of emotional sensitivity. This work forms the first study to demonstrate the stability of fNIRS based dynamic FC states as a biomarker. In conclusion, the results support that the state distribution of dynamic FC could help reveal the differentiating factors between the nursing students and registered nurses during decision making, and it is anticipated that the biomarkers might be used as indicators when developing professional training related to emotional sensitivity.

Coordinated Allocation of BESS and SOP in High PV Penetrated Distribution Network Incorporating DR and CVR Schemes

This article investigates the coordinated allocation of battery energy storage system (BESS) and soft open point (SOP) in high photovoltaic penetrated distribution network incorporating demand response (DR), and conservation voltage reduction (CVR) schemes. To do so, a new two-stage coordinated optimization framework has been developed for integrated planning of BESS and SOP considering DR and CVR schemes. In stage 1, optimal location and size determination of BESS and SOP devices has been carried out simultaneously. Whereas, stage 2 performs the optimal operation of BESS and SOP corresponding to CVR and DR schemes. The objective of the proposed methodology is to minimize the total investment and operating cost of BESS and SOP devices. Besides, it includes the cost of purchased power from substation, cost of energy not served and cost of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\text{CO}_2$</tex-math></inline-formula> emission. Meanwhile, to address high-level uncertainties related to PV generation and load demands, a stochastic module has been adopted, which produces several scenarios by Monte Carlo simulation at each hour. Furthermore, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> -means clustering technique has been employed to determine the reduced scenarios with high quality and diversity. The proposed framework has been implemented on IEEE 33-bus distribution system for different cases and solved by using hybrid optimization solver. The test results demonstrate the effectiveness of proposed framework in improving the system efficiency, enhancing the flexibility, reliability and reducing carbon emission footprint of distribution systems.

Multi-Domain Feature Extraction for Improved Clustering of Smart Meter Data

The advent of smart grid is a revolution that has enabled power distribution in a more efficient way. However, load forecasting, demand response management and accurate consumer load profiling using smart meter data continue to be challenging industry and research problems. Clustering is an efficient technique for load profiling. K-means clustering algorithm for clustering electricity consumers based on raw meter data directly result in cumbersome, redundant and inefficient computations. This paper presents a methodology for reducing the raw data set dimension via features extraction and cluster the load profiles based on computed features. The feature set formed comprises of Singular Values by Singular Value Decomposition and Wavelet Energy Entropy of approximate and detailed Coefficients. K means Clustering technique is used. The proposed method enables efficient and quick clustering and at the same time the information content in load profiling is preserved. The time consumed for clustering of feature set formed is found to be much less than that of raw data set. By comparing the Silhouette Values K = 6 was found to be the optimal number of clusters with average silhouette coefficient around 0.79. Clustering of load profiles both for Raw Data Set as well as computed Feature Set are compared by evaluating average silhouette value, number of negative silhouettes and computation time for clustering and Silhouette Coefficient was found to be 0.79 by proposed methodology showing better clustering result as compared to raw dataset.

Brain Tumor Segmentation and Classification using Hybrid clustering technique and SVM Classifier

Brain tumor Detection is a primary concern in today’s life. So a computer aided technology must be implemented for an accurate detection and identification of brain tumor. The tumor can be detected using various classification techniques from brain MR Images. In this paper segmentation process is being done using K means Clustering technique and Binary Thresholding, the features from the images are then extracted using GLCM where six texture features are extracted and SVM Classifier is being used for classification of the images. This proposed method shows an accuracy of 97.12%.

Performance evaluation of cybercriminal detection through cluster computing techniques

The cybercrime has evolved as the years ago and the criminals use different techniques to capture the victims. Researchers have devised various techniques to detect the cybercriminal. Hence, a new technology has to be used to detect the Cybercrime. From the literature, it is analysed that many researchers have used various techniques to detect the Cybercrime. In this paper, the performance evaluation of various cybercriminal detection techniques are analysed using cluster computing techniques and Matlab is used to determine their performance. In this paper, 25 attributes are analysed to detect the cybercriminal through various techniques such as Gaussian Clustering, K Means, Fuzzy C Means and Fuzzy Clustering. Some of the attributes are taken as varying attributes and some of the attributes are taken as the nonvarying attributes. The crime clusters and genuine clusters are determined. The reasons for choosing the various attributes and the various techniques are also given. The analysis is further strengthened by changing the number of attributes used in each technique i.e., 25 attributes, 15 attributes, 10 attributes and 5 attributes. The analysis is further continued by using ¾ of total attributes, ½ of total attributes and ¼ of total attributes. The efficiency and time complexity is determined. In the paper justification of a number of attributes selected in each technique is also given. The accuracy analysis is done by comparing with Gaussian Clustering with K Means Clustering technique, Fuzzy C Means, and user profile analysis. The attributes which are significantly contributing for identifying the criminals are determined in each technique.

Use of Clustering to Reduce the Number of Different Setting Groups for Adaptive Coordination of Overcurrent Relays

The facility to save multiple setting groups (SGs) within digital overcurrent relays (OCRs) is to adapt the active setting of the relays to the current topology of the power network. However, the number of available SGs is much lower than the possible topologies of the network, so this facility is usually applied to a few local changes in the network. This paper aims to extend the application to a complete set of topologies. A proper index is introduced and applied with a k -means clustering technique to classify the topologies into some clusters, whose number is equal to the number of SGs. Then, for every topology cluster, the optimal settings are calculated and saved in the OCRs as a distinct SG to be activated by changing the topology to any member of the cluster. Applying this technique to OCRs of the IEEE 14-bus test network and a 63/20kV substation verifies its effectiveness on reserving coordination for all topologies and reducing the relays’ time dial setting. Using available communication links, this technique can be implemented by making a computer network between substations of the power network without needing to interact with the control center.

Probabilistic Power Flow Analysis of Power Systems Incorporating Tidal Current Generation

A probabilistic power-flow analysis method for power systems with tidal power sources is presented in this paper. The regularity of tidal power is modeled using a k -means clustering technique and the randomness of tidal power is modeled by using a nonparametric kernel density estimation method. A stochastic sampling method is also developed to generate random samples of tidal power time series for Monte Carlo based probabilistic power-flow analysis. The influence of tidal current generation on power flows is then evaluated and quantified considering both the regularity and randomness of tidal power. The measured tidal current speed data of two different locations in Florida and Alaska states, USA, and the IEEE 57-bus standard test system are used to verify the correctness and effectiveness of the presented probabilistic power-flow analysis method.

Reliability and Cost Analysis of Multistage Boost Converters Connected to PV Panels

In photovoltaic (PV) generation systems, dc–dc boost converters are responsible for maximum power point tracking and voltage regulation. The intent of this paper is to determine the number of the stages in an interleaved boost converter interfacing PV panels for achieving a reliable and costly optimized structure. A comparative study has been done on different modes of operation. including redundant operation or parallel operation in a two-stage interleaved converter. The comparison indicates that working in simultaneous mode would be more reliable. Contemplating this fact, reliability equation of a three-stage interleaved converter is calculated for simultaneous mode of operation. Considering dark hours of the day, a k -means clustering technique has been utilized in the reliability calculations based on the output data of a PV generation system installed at the campus. Besides, mean time to failure criterion is considered in the reliability analysis. Simulation and experimental results are analyzed, considering the costs. The results determine the optimum structure for interfacing the PV panels to the grid.

Superpixel Optimization Using Higher Order Energy

A novel superpixel extraction algorithm using a higher order energy optimization framework is proposed in this paper. We first adopt the $k$ -means clustering technique to quickly get an initial superpixel result. Then a higher order energy function is employed to optimize and refine these initial superpixels. We use a more general higher order energy function that includes a first-order data term, a second-order smoothness term, and a higher order term. The presegments are employed to provide the prior information of sufficient edges and segment regions for our higher order energy term. According to the texture measurement in different local regions, our algorithm adaptively computes the proper ratios of different energy terms to obtain a better superpixel performance. The experimental results demonstrate that our method using the higher order energy generates better results with well-aligned boundaries and homogeneous effects than the existing superpixel algorithms.

Double-Layer Fuzzy Fusion for Multiview Through-Wall Radar Images

In this letter, a double-layer fuzzy fusion method is proposed to produce a higher quality image by combing multiple through-wall radar images of the same scene. The first layer is termed as intraimage fusion, in which the global and local degrees of membership are fused within each input image based on intraimage fuzzy rules. Then, in the second layer, interimage fusion is performed between input images using interimage fuzzy rules, and the output image is obtained. Compared with other fusion methods, the proposed method decreases the possibility of losing a weak target and suppresses the clutter by taking into account the local information while enhancing the target intensity values. The method also introduces a $K$ -means clustering technique to automatically select the parameters of the membership functions. The performance of the proposed method is evaluated using both simulated and real experimental data.

Genetic Algorithm Based on K-means-Clustering Technique for Multi-objective Resource Allocation Problems

This paper presents genetic algorithm based on K - means clustering technique for solving multi objective resource allocation problem (MORAP). By using k - means clustering technique, population can be divided int o a specific number of subpopulations with dynamic size. In this way, different GA operators (crossover and mutation) can be applied to each subpopulation instead of one GA operators applied to the whole population. The purpose of implementing K - means clus tering technique is preserving and introducing diversity. Also it enable the algorithm to avoid local minima by preventing the population of chromosomes from becoming too similar to each other. Two test problems taken from the literature are used to compar e the performance of the proposed approach with the competing algorithms. The results have been demonstrated the superiority of the proposed algorithm and its capability to solve MORAP.

Regional Frequency Analysis of Droughts in China: A Multivariate Perspective

Joint probability behavior of droughts is important for China due to the fact that China is the agricultural country with the largest population in the world and it is particularly the case in the backdrop of intensifying weather extremes in a warming climate. In this case, regionalization of droughts is done using Fuzzy C- Means (FCM) clustering technique and also multivariate L-moment method. Besides, copula is used to estimate regional joint probability in terms of drought duration and severity. Evaluation of uncertainty in the joint probability curves is done using the Bootstrap resampling technique. The results indicate that: (1) five homogenous regions of droughts are subdivided. Regionalization in this study clarified the changing properties or nature of droughts, i.e., the blurred or ambiguous boundaries of the drought-impacted regions; (2) droughts in the northwest China are characterized by longer drought duration and larger drought severity, and the occurrence of the droughts in the northwest China is subject to be higher due to longer waiting time between drought events. Adverse is found for changes of droughts in the southeast China. The droughts in the north China are moderate in terms of drought duration and severity and also waiting time between drought events when compared to those in the northwest and southeast China; (3) the regional joint frequency curves are obtained with respect to drought duration and severity using the bivariate copula functions. Then the joint probabilities of droughts can be calculated using the regional probability curves and also results of mean drought duration, drought severity and waiting time between drought events. Furthermore, droughts in the regions without meteorological data can also be estimated in terms of joint probability using index-drought method proposed in this study. This study will provides theoretical and practical grounds for development and enhancement of human mitigation to drought hazards in China, and is of great importance in terms of planning and management of water resources and agricultural activities in the backdrop of intensifying weather extremes under the influences of warming climate.