Optimization Clustering Algorithm Research Articles

DACOR: A distributed ACO‐based routing protocol for mitigating the hot spot problem in fog‐enabled WSN architecture

SummaryAnt colony optimization (ACO) and unequal clustering algorithms in wireless sensor networks (WSNs) prove their efficiency in protracting the network lifetime. However, the existing ACO and unequal clustering algorithms, respectively, do not consider jointly energy efficiency and reliability and focus only on some normal parameters to adjust the cluster radius, then neglecting the cluster head (CH) neighborhood information as it is wise to reduce the cluster radius when there are more neighbor CHs in order to balance the load and energy consumption. To resolve these problems, we propose a fault‐tolerant distributed ACO‐based routing (DACOR) protocol for mitigating the hot spot problem in fog‐enabled WSN architecture. To improve the performance of the network, we propose a multiple fog nodes (FNs) and unequal clustering‐based network model. The proposed model is energy efficient as it avoids repetitive clustering and affects CHs to FNs based on distance. Also, unlike the existing works which use either single FN/sink‐based unequal clustering or multiple FNs/sinks to mitigate hot spot problem, we propose to distribute unequal clustering to multiple FNs (partitions). Additionally, we formulate a different rule to calculate the cluster radius based on significant parameters ensuring energy efficiency and balancing. To route data from source to destination, we devise a new probabilistic formula which considers not only energy efficiency but also reliability. The performance of the proposed DACOR protocol has been investigated under different scenarios through simulations. The results show that the proposed DACOR protocol outperforms the existing protocols in terms of various main metrics.

Evolutionary state‐based novel multi‐objective periodic bacterial foraging optimization algorithm for data clustering

AbstractClustering divides objects into groups based on similarity. However, traditional clustering approaches are plagued by their difficulty in dealing with data with complex structure and high dimensionality, as well as their inability in solving multi‐objective data clustering problems. To address these issues, an evolutionary state‐based novel multi‐objective periodic bacterial foraging optimization algorithm (ES‐NMPBFO) is proposed in this article. The algorithm is designed to alleviate the high‐computing complexity of the standard bacterial foraging optimization (BFO) algorithm by introducing periodic BFO. Moreover, two learning strategies, global best individual (gbest) and personal historical best individual (pbest), are used in the chemotaxis operation to enhance the convergence speed and guide the bacteria to the optimum position. Two elimination‐dispersal operations are also proposed to prevent falling into local optima and improve the diversity of solutions. The proposed algorithm is compared with five other algorithms on six validity indexes in two data clustering cases comprising nine general benchmark datasets and four credit risk assessment datasets. The experimental results suggest that the proposed algorithm significantly outperforms the competing approaches. To further examine the effectiveness of the proposed strategies, two variants of ES‐NMPBFO were designed, and all three forms of ES‐NMPBFO were tested. The experimental results show that all of the proposed strategies are conducive to the improvement of solution quality, diversity and convergence.

Neural Network Data Mining Clustering Optimization Algorithm

Data mining technology is an effective way to solve the problem of rich data and poor knowledge. Cluster analysis is an important content of data mining, including analysis ideas based on partition, hierarchy, density, grid and model. Neural network clustering is a typical clustering method based on model thinking. It is an organic combination of brain cognitive science and data mining. It has a strong theoretical connection with actual brain processing knowledge. This paper uses the K-means algorithm to optimize the neural network clustering data mining algorithm, and designs experiments to verify the neural network data mining clustering optimization algorithm proposed in this paper. The experimental research results in this paper show that on four UCI datasets, the mean MP values of the neural network data mining clustering optimization algorithms are 77.9 and 87.72, respectively, which are greater than the values of the other two algorithms. This paper also applies the algorithm to the study of the distribution of remaining oil. The algorithm has achieved obvious results in the cluster analysis of the degree of flooding.

Optimal Cluster Number Identification using Ant Colony Optimization

Due to the hopeful application of gathering information from unreachable position, wireless sensor network creates an immense challenge for data routing to maximize the communication with more energy efficient. In order to design the energy efficient routing, the optimization based clustering protocols are more preferred in wireless sensor network. In this paper, we have proposed competent optimization based algorithm called ANT colony Optimization clustering algorithm for creating the energy efficient routing path. WSNs comprise of countless spatially dispersed sensor hubs associated through the remote medium to screen and record the actual data from the climate. The hubs of WSN are battery controlled, so after a specific period it free whole energy. This energy limitation influences the lifetime of the organization. The goal of this review is to limit the generally energy utilization and to boost the organization lifetime. In this paper, a new method to optimize number of clusters is presented with suitable results..

Local neighbour spider monkey optimization algorithm for data clustering

Local neighbour spider monkey optimization algorithm for data clustering

Optimized Clustering Algorithm for Comparative Analysis of Different Prenatal Corticosteroid Neurological Deficits in Premature Infants through Magnetic Reasoning Imaging (MRI).

Objective This study aimed to explore the application of different prenatal corticosteroids in the assessment of neurological deficits and prognosis in premature infants through Magnetic Reasoning Imaging (MRI) under optimized cluster algorithm. Methods 100 pregnant women with threatened preterm labor were retrospectively analyzed, in which 38 pregnant women with lasting threatened preterm labor (group A) were treated with multiple courses of antenatal corticosteroids (dexamethasone treatment) and 62 cases of pregnant women with threatened preterm labor (group B) were treated with single course of dexamethasone treatment. Craniocerebral MRI images based on optimal clustering algorithm were used to examine neonates. Neonatal hypoxic-ischemic encephalopathy (HIE) rate, serum neuron-specific enolase (NSE) concentration, neonatal behavioral neurological score (NBNA), respiratory distress syndrome (RDS) rate, perinatal mortality, neonatal birth weight, and maternal complications rate of two groups were compared. Results Compared with other traditional image segmentation algorithms, this algorithm had the best segmentation effect, the shortest running time (1.43 s), the least number of iterations (5 times), and the highest segmentation accuracy (97.98%). There was no significant difference in the HIE rate, serum NSE concentration, NBNA score, RDS score, and perinatal mortality in group A and group B (P > 0.05). Compared with group B, neonates' body weight in group A was decreased, while the maternal complication rate in group A was increased (P < 0.05). Conclusion MRI images based on optimized clustering algorithm can be used in the diagnosis of neonatal hypoxic-ischemic encephalopathy. There is no significant difference in the application of different antenatal corticosteroids affecting premature nerve function defect and prognosis, but multiple courses of antenatal corticosteroids can affect neonatal body mass and increased maternal complications to a certain extent; therefore, before threatened premature delivery treatment, the pros and cons of multiple courses of antenatal corticosteroids should fully be considered and in the treatment, measures should be actively taken to alleviate the side effect.

Public Housing Allocation Model in the Guangdong-Hong Kong-Macao Greater Bay Area under Clustering Algorithm

In the Guangdong-Hong Kong-Macao Greater Bay Area (Bay Area), the allocation methods of public rental housing are analyzed to achieve scientific and fair housing allocation as much as possible, so as to protect the housing demand of low-income and middle-income families. The housing model in the Bay Area is analyzed firstly, and the key points of public rental housing and allocation management models are discussed comprehensively. Furthermore, a method based on rough-based fuzzy clustering (RFC) is proposed to analyze the housing demands of security groups, and a public housing allocation model is constructed based on actual demand of residents. The housing allocation plan is given and decided by the decision-making department based on the demand of the security objects and the characteristics of public housing. The simulation experiments are performed on the clustering algorithm optimized based on rough set feature selection. On the Chess data set, the optimized clustering algorithm shows an obvious improvement in clustering accuracy and recall rate compared with the traditional clustering algorithms, which are 0.76 and 0.95, respectively. The bilateral matching method based on fuzzy axiom design can fully consider the actual needs of both the supply and demand of the housing security, which is beneficial to improve the rationality and correctness of public housing allocation. The allocation method of public housing based on demand clustering analysis focuses on improving the housing security level and strives to meet the higher-level housing improvement needs of housing security objects, so as to provide security objects with more expected living conditions and improve housing allocation effect.

Early Detection of Product Reliability Based on the Parameters of the Production Line and Warranty Data

Reliability is one of the key dimensions of the quality of services and products that should be always evaluated. Growth and development of industries can be achieved by appropriate reliability engineering of products. Companies should evaluate and predict the reliability of products and accordingly find and fix the potential problems. In this regard, early detection of reliability problems based on the parameters of the production line or quality test results can prevent future warranty costs. Early detection of reliability problems based on production process and test data has not gained much attention in the literature. Therefore, an early detection model for predicting the reliability of products according to their quality test results is proposed in this research. For this purpose, hot test and warranty data of car engines manufactured by an automotive company are utilized. This data are prepared to predict engine reliability after preprocessing and removing inefficient data. Then, engines are divided into two homogeneous clusters using particle swarm optimization (PSO) clustering algorithm. Afterwards, the data in these clusters are used to feed the Artificial Neural Network (ANN) to predict the reliability of the engines. The obtained results show that the proposed ANN-based method is able to predict the reliability of the engines based on engine kilometers operated and hot test results. Also, it is shown that the proposed method outperforms the Cox proportional hazards model which has previously been used for early detection of product reliability.

Improved Self-Adaptive ACS Algorithm to Determine the Optimal Number of Clusters

A fundamental problem in data clustering is how to determine the correct number of clusters. The k-adaptive medoid set ant colony optimization (ACO) clustering (METACOC-K) algorithm is superior in solving clustering problems. However, METACOC-K does not guarantee in finding the best number of clusters. It assumed the number of clusters based on an adaptive parameter strategy that lacks feedback learning. This has restrained the algorithm in producing compact clusters and the optimal number of clusters. In this paper, a self-adaptive ACO clustering (S-ACOC) algorithm is proposed to produce the optimal number of clusters by incorporating a self-adaptive parameter strategy. The S-ACOC algorithm is a centroid-based algorithm that automatically adjusts the number of clusters during the algorithm run. The selection of the number of clusters is based on a construction graph that reflects the influence of a pheromone in algorithm learning. Experiments were conducted on real-world datasets to evaluate the performance of the proposed algorithm. The external evaluation metrics (purity, F-measure, and entropy) were used to compare the results of the proposed algorithm with other swarm clustering algorithms, including a genetic algorithm (GA), particle swarm optimization (PSO), and METACOC-K. Results showed that S-ACOC provides higher purity (50%) and lower entropy (40%) than GA, PSO, and METACOC-K. Experiments were also performed on several predefined clusters, and results demonstrate that the S-ACOC algorithm is superior to GA, PSO, and METACOC-K. Based on the superior performance, S-ACOC can be used to solve clustering problems in various application domains.

Forecasting retail sales based on cheng fuzzy time series and particle swarm optimization clustering algorithm

Use of the conventional forecasting method, which is based on trend data with average sales in the last few months, results inaccurate forecasting due to a large difference in data, this is the same as fuzzy forecasting with the same interval length or static. Therefore, this paper recommends using the Cheng forecasting method combined with the Particle Swarm Optimization (PSO) algorithm. We use an artificial intelligence, i.e., PSO algorithm to set non-static length of intervals each cluster on Cheng method. The comparison of this method yields a better root mean square error (RMSE) value for each cluster on the recommended method.

Study of cell clustering optimization algorithm for load balancing in 5G scenario

In the context of 5G networks at this stage, massive data transmission and highly dense deployment of base stations make severe interference and huge load volumes an inevitable new challenge. The traditional interference coordinated cell-clustering scheme only considers the inter-micro base station path loss without jointly considering load balancing, an important factor affecting system performance. This paper first introduces the existing cell-clustering schemes, and then proposes a cell-clustering optimization algorithm that combines path loss and load balancing on the basis of 5G cellular network architecture, and then simulate the optimization algorithm and compare the results with the traditional schemes to verify its ability to balance the load, increase the amount of transceiver data, and optimize the resource allocation and performance effects.

An Ensemble of Community Detection in Social Networks Using Clustering of Users Demographic and Topological Information

Background: One of the great challenges in social network analysis is community detection. Community is a group of users which have high intra connections and sparse inter connections. Community detection or clustering reveals community structure of social networks and hidden relationships among their constituents. Nowadays, many different methods are proposed to detect community structures in social networks from different perspectives, but none of them can be a constant winner. Therefore, an ensemble of different methods can potentially improve the final result. Methods: In this paper, we present a framework for different methods to be combined for community detection. This method is a combination of genetic algorithms, particle swarm optimization, kmeans clustering and Louvain clustering algorithms. Our method uses topological and demographic information to identify communities and can automatically determine the number of optimal communities. Results and Conclusion: Quantitative evaluations based on extensive experiments on Ego- Facebook social network dataset reveals that the method presented in this study achieves favorable results, which are quite superior to other relevant algorithms in the literature. • Discovering relationships between individuals by analyzing social networks. • Providing identifying communities algorithms based on different clustering methods. • An ensemble of community detection consisting of GA, PSO, k-means and Louvain clustering. • The proposed method is better than the TSA method at silhouette and modularity criterion. Demographic information also relates to the profile of users and their shared tweets.

Fast Optimal Circular Clustering and Applications on Round Genomes.

Round genomes are found in bacteria, plant chloroplasts, and mitochondria. Genetic or epigenetic marks can present biologically interesting clusters along a circular genome. The circular data clustering problem groups N points on a circle into K clusters to minimize the within-cluster sum of squared distances. Repeatedly applying the K-means algorithm takes quadratic time, impractical for large circular datasets. To overcome this issue, we developed a reproducible fast optimal circular clustering (FOCC) algorithm of worst-case O(KN log2 N) time. The core is a fast optimal framed clustering algorithm, which we designed by integrating two divide-and-conquer and one bracket dynamic programming strategies. The algorithm is optimal based on a property of monotonic increasing cluster borders over frames on linearized data. On clustering 50,000 circular data points, FOCC outruns brute-force or heuristic circular clustering by three orders of magnitude in time. We produced clusters of CpG sites and genes along three round genomes, exhibiting higher quality than heuristic clustering. More broadly, the presented subquadratic-time algorithms offer the fastest known solution to not only framed and circular clustering, but also angular, periodical, and looped clustering. We implemented these algorithms in the R package 'OptCirClust' (https://CRAN.R-project.org/package=OptCirClust).

SLICACO: An automated novel hybrid approach for dermatoscopic melanocytic skin lesion segmentation

AbstractLow contrast images and blurriness pose challenge in the over‐segmentation of image, which increases model complexities. In this work, a novel hybrid dermoscopic skin‐lesion segmentation method, namely SLICACO, is proposed incorporating the simple linear iterative clustering (SLIC) and ant colony optimization (ACO) algorithms. The working of proposed method is multifold. First, over‐segmentation of preprocessed image is generated using SLIC super‐pixel technique. Second, clusters of super‐pixels generated by SLIC are used by ACO with the pixels of similar intensity for edge detection and seek for the optimum pathway in a strained zone. Third, lesion area is segmented using the Convex Hull and Thresholding. Fourth, Erosion Filtering is used to obtain the final segmented image. The performance of SLICACO is assessed on five benchmark dermatoscopic datasets and compared with deep learning models to test its generalizing behavior. Promising results are obtained on the PH2 archive data set with an accuracy of 95.9%.

An intelligent cluster optimization algorithm based on Whale Optimization Algorithm for VANETs (WOACNET).

Vehicular Ad hoc Networks (VANETs) an important category in networking focuses on many applications, such as safety and intelligent traffic management systems. The high node mobility and sparse vehicle distribution (on the road) compromise VANETs network scalability and rapid topology, hence creating major challenges, such as network physical layout formation, unstable links to enable robust, reliable, and scalable vehicle communication, especially in a dense traffic network. This study discusses a novel optimization approach considering transmission range, node density, speed, direction, and grid size during clustering. Whale Optimization Algorithm for Clustering in Vehicular Ad hoc Networks (WOACNET) was introduced to select an optimum cluster head (CH) and was calculated and evaluated based on intelligence and capability. Initially, simulations were performed, Subsequently, rigorous experimentations were conducted on WOACNET. The model was compared and evaluated with state-of-the-art well-established other methods, such as Gray Wolf Optimization (GWO) and Ant Lion Optimization (ALO) employing various performance metrics. The results demonstrate that the developed method performance is well ahead compared to other methods in VANET in terms of cluster head, varying transmission ranges, grid size, and nodes. The developed method results in achieving an overall 46% enhancement in cluster optimization and an F-value of 31.64 compared to other established methods (11.95 and 22.50) consequently, increase in cluster lifetime.

Application of the novel harmony search optimization algorithm for DBSCAN clustering

At present, the DBSCAN clustering algorithm has been commonly used principally due to its ability in discovering clusters with arbitrary shapes. When the cluster number K is predefined, though the partitional clustering methods can perform efficiently, they cannot process the non-convex clustering and easily fall into local optimum. Thereby the concept of K-DBSCAN clustering is proposed in this paper. But the basic DBSCAN has a crucial defect, that is, difficult to predict the suitable clustering parameters. Here, the well-known harmony search (HS) optimization algorithm is considered to deal with this problem. By modifying the original HS, the novel harmony search (novel-HS) is put forward, which can improve the accuracy of results as well as enhance the robustness of optimization. In K-DBSCAN, the novel-HS is used to optimize the clustering parameters of DBSCAN to obtain better clustering effect with the number of K classifications. Experimental results show that the designed clustering method has superior performance to others and can be successfully considered as a new clustering scheme for further research.

Mushroom Edibility Identification Applying CBR and Ant Lion Techniques in Multi-sensor Environment

Wireless sensor networks (WSN) are part of our daily life as they play a vital role in applications of various domains. Energy optimization is a major challenge of WSN as they are operated through the battery. Data exchange is a significant and often performed operation in any WSN which has to be operated with less energy consumption. Bio-inspired clustering protocols are proving a success in minimizing energy consumption and are of current research interest by many researchers. This paper briefs on work carried on the classification of mushroom into edible or non-edible. Since most mushrooms are dangerous to health and may lead to death, henceforth is it is essential to identify the edibility of mushroom. Mushroom features identified by sensor such as ring, odur, spore_print_color, stalk_color_above, stalk_surface_above, and gill size are forwarded to base station (BS). Ant Lion optimization clustering algorithm (ALOC) is adopted in the routing of information to BS. ALOC avoids improper clustering reducing multiple messages at BS. The work is divided into two phases in the initial phase sensing of mushroom features and forwarding to BS is performed through WSN. In the second phase, the decision is taken whether mushroom is edible or not applying class-based association rules (CBA). The simulation results show ALOC is better than LEACH, ABC PSO and MLEACH through evaluation results in terms of network lifetime, energy consumption and retained alive nodes. correlation-based feature selection (CFS) with three filter search techniques: genetic algorithm (GA), evolutionary algorithm (EA), and particle swarm optimization (PSO) at BS as an evaluation method for selection of significant feature selection. CBA rules were generated using these subsets of significant features; hence resulted in a limited number of strong rules which are reliable and sufficient enough to classify the mushroom as edible or not. The patterns and rules generated using the proposed approach avoid the generation of duplicate and irrelevant rules and henceforth simplifies the analysis process using a reliable and interesting set of rules.

A hybrid approach for text document clustering using Jaya optimization algorithm

In this digital era, millions of Internet users are contributing vast amounts of data in the form of unstructured text documents. Organizing this material is a tedious task. The clustering of text document plays a vital role for organizing these unstructured text documents. In our paper, we make use of Hybrid Jaya Optimization algorithm (HJO) for text Document Clustering (DC), referred to as HJO-DC. We have used the Silhouette index as a metric to measure the quality of a solution. The proposed work is compared with partitioning techniques such as K-Means and K-Medoids and metaheuristic techniques such as Genetic algorithm, Cuckoo Search, Particle Swarm Optimizer, Firefly and Grey Wolf Optimizer. Remarkably, the proposed algorithm achieves the highest quality clustering in all benchmark examples.

Data Analysis Using Representation Theory and Clustering Algorithms

This work aims to expand the knowledge of the area of data analysis through both persistence homology, as well as representations of directed graphs. To be specific, we looked for how we can analyze homology cluster groups using agglomerative Hierarchical Clustering algorithms and methods. Additionally, the Wine data, which is offered in R studio, was analyzed using various cluster algorithms such as Hierarchical Clustering, K-Means Clustering, and PAM Clustering. The goal of the analysis was to find out which cluster's method is proper for a given numerical data set. By testing the data, we tried to find the agglomerative hierarchical clustering method that will be the optimal clustering algorithm among these three; K-Means, PAM, and Random Forest methods. By comparing each model's accuracy value with cultivar coefficients, we came with a conclusion that K-Means methods are the most helpful when working with numerical variables. On the other hand, PAM clustering and Gower with random forest are the most beneficial approaches when working with categorical variables. All these tests can determine the optimal number of clustering groups, given the data set, and by doing the proper analysis. Using those the project, we can apply our method to several industrial areas such that clinical, business, and others. For example, people can make different groups based on each patient who has a common disease, required therapy, and other things in the clinical society. Additionally, for the business area, people can expect to get several clustered groups based on the marginal profit, marginal cost, or other economic indicators.

A hybridization of SKH and RKFCM clustering optimization algorithm for efficient moving object exploration

Object detection is a really crucial application of image processing. It is of essential value for object active surveillance and various other applications. Hence, the object detection has been extensively investigated. Refined Kernel fuzzy c-means system still carries a couple of downsides, for instance, to reduce the convergence level, obtaining stuck in the regional area minima and problem to initiation level of sensitiveness. To conquer the over problems, the following suggested strategy for stud krill herd Clustering Optimization Algorithm. This paper stands for an optimizing approach to global optimizing utilizing a unique variation of KH (Krill Herd). This approach is termed as Stud Krill Herd (SKH).The stud krill herd Clustering Optimization procedure utilizes to find the optimal centroid. At first, the background and foreground area partition is finished by hybridization of refined kernel fuzzy c means algorithm (RKFCM) with stud krill herd Clustering Optimization procedure. The suggested new strategy is scholarly and furthermore dynamic clustering system for dividing the moving object. This research study performs recommended a reliable object detection making use of hybridization of stud krill herd Clustering optimizing and RKFCM. Moving object tracing is done via the blob detection which happens under the tracing phase. The assessment phase has characteristic abstraction and categorization. High and appearance-based and quality based attributes are mined from fine-tuned frames whichever attended to classification. Considering that categorization we are developing usage of J48 (C4.5) i.e., decision tree based classification. The effectiveness of the advised method is analyzed through prior approaches k-NN and MLP in regard to accuracy, f-measure, ROC and recall.