Self-Optimal Clustering Technique Using Optimized Threshold Function

The damaged areas of brain tissues can be extracted by using segmentation methods, most of which are based on the integration of machine learning and data mining techniques. An important segmentation method is to utilize clustering techniques, especially the fuzzy C-means (FCM) clustering technique, which is sufficiently accurate and not overly sensitive to imaging noise. Therefore, the FCM technique is appropriate for multiple sclerosis diagnosis, although the optimal selection of cluster centers can affect segmentation. They are difficult to select because this is an NP-hard problem. In this study, the Harris Hawks optimization (HHO) algorithm was used for the optimal selection of cluster centers in segmentation and FCM algorithms. The HHO is more accurate than other conventional algorithms such as the genetic algorithm and particle swarm optimization. In the proposed method, every membership matrix is assumed as a hawk or an HHO member. The next step is to generate a population of hawks or membership matrices, the most optimal of which is selected to find the optimal cluster centers to decrease the multiple sclerosis clustering error. According to the tests conducted on a number of brain MRIs, the proposed method outperformed the FCM clustering and other techniques such as the k-NN algorithm, support vector machine, and hybrid data mining methods in accuracy.

This paper presents a hard clustering technique using fireworks algorithm with adaptive transfer function (FWAATF) for image segmentation. The fireworks algorithm (FWA) is a recently developed new Swarm Intelligence (SI) algorithm for function optimization. This algorithm simulates the process of fireworks explosion in the night sky. The main characteristic of FWA is the good balance between exploration and exploitation during the search process. The exploitation is done using good fireworks whereas the bad fireworks are responsible for exploration. FWA shows its efficiency and effectiveness in numerical function optimization over other SI algorithm like particle swarm optimization (PSO). FWA-ATF is a modified version of basic FWA and in this work, it is used in hard clustering technique to segment the image. FWA-ATF is used to find the optimal cluster centroids corresponding to different regions in the image. The proposed clustering technique is applied to segment four benchmark images and the well-known cluster validity index-Dunn's Index is used to measure the performance of the proposed clustering technique quantitatively. The performance of the proposed method is compared with clustering using K-means, PSO and basic FWA. The experimental results demonstrates that the proposed clustering technique with FWA-ATF performs better than other methods in segmentation for most of the images.

A systematic approach for detecting and clustering distributed cyber scanning

Abstract. Long-term measurements of particle number size distribution (PNSD) produce a very large number of observations and their analysis requires an efficient approach in order to produce results in the least possible time and with maximum accuracy. Clustering techniques are a family of sophisticated methods that have been recently employed to analyse PNSD data; however, very little information is available comparing the performance of different clustering techniques on PNSD data. This study aims to apply several clustering techniques (i.e. K means, PAM, CLARA and SOM) to PNSD data, in order to identify and apply the optimum technique to PNSD data measured at 25 sites across Brisbane, Australia. A new method, based on the Generalised Additive Model (GAM) with a basis of penalised B-splines, was proposed to parameterise the PNSD data and the temporal weight of each cluster was also estimated using the GAM. In addition, each cluster was associated with its possible source based on the results of this parameterisation, together with the characteristics of each cluster. The performances of four clustering techniques were compared using the Dunn index and Silhouette width validation values and the K means technique was found to have the highest performance, with five clusters being the optimum. Therefore, five clusters were found within the data using the K means technique. The diurnal occurrence of each cluster was used together with other air quality parameters, temporal trends and the physical properties of each cluster, in order to attribute each cluster to its source and origin. The five clusters were attributed to three major sources and origins, including regional background particles, photochemically induced nucleated particles and vehicle generated particles. Overall, clustering was found to be an effective technique for attributing each particle size spectrum to its source and the GAM was suitable to parameterise the PNSD data. These two techniques can help researchers immensely in analysing PNSD data for characterisation and source apportionment purposes.

Conventional clustering techniques like FCM, K-Means, Mountain clustering etc. face the main problem of excessive data while dealing with the very big size images. Due to higher order dependency of clustering techniques on the number of data points, time complexity increases excessively while dealing with very large size images. This paper proposes an advanced version of mountain clustering technique, Fast Mountain clustering (FMC), for segmentation of grayscale images whose run time is almost independent of size of image. The proposed approach consists of defining the dataset in another domain which makes the clustering almost independent of size of the data. The obtained results are compared with the widely used techniques like FCM, K-Means, IMC and found out to be better on the basis of cluster validity measures Global silhouette index (GS) and Partition Index (SC).

Any data in the real world can be organized properly by Clustering the data using some clustering techniques and in these clustering techniques Fuzzy C-Means (FCM) is a very recent and better technique that can mold the data with good logic and in a highly accurate manner. FCM is same as K-Means clustering technique but FCM is developed with some Fuzzy means. FCM is associated with some constraints as FCM is more responsive as on the order how the clusters were arranged in the beginning of applying the technique. Even though it has its own defects when dealing with large data, it cannot get the finest solution as the data need some optimization? So to get some better accuracy, In our proposed paper we are applying an enhanced FCM for clustering and for obtaining precise results we are involving GSA based on gravitation laws and perception of masses. This optimizing technique (GSA) can be used to improve the limitations in order to gain a well-defined system performance; GSA can efficiently deal with many propositions of large data. Here we are going to develop a Map-Reduce mechanism to effectively deal with large data. The expected result can be achieved by choosing the better finest candidates and grouping them in the reduction mode to get the finest solution. Evaluation of practical results will be optimizing when compared to other existing similar techniques.

Recently, indoor localisation techniques that use wireless local area network (WLAN) be a-con signals have gained much attention by the research communities. Many localisations methods are used to estimate the user of mobile device in indoor environments. However, the accuracy of these methods is affected by the nature of the test-bed environment. In this paper, we introduce an experimental test-bed in a typical indoor environment. We used a finger printing-based localisation algorithm to estimate the user location. The fingerprinting technique consists of two phases: offline phase and online phase. In the offline phase, calibration points are collected at certain places in floor to build a radio map. In the online phase, deterministic and probabilistic approaches are applied in order to get the correct estimated position of a mobile device. In deterministic approach, the position of mobile device estimated by K-nearest neighbour (KNN). In probabilistic approach, the position of mobile device estimated by Bayesian network (BN). Clustering technique is proposed to improve the system's accuracy and reduce the radio map size in the offline phase. We present experimental results that improved the system accuracy and reduce the size of radio map by using the proposed clustering technique.

Performance analysis of clustering techniques over microarray data: A case study

Urban transportation systems encounter significant challenges in extracting meaningful traffic patterns from extensive historical datasets, a critical aspect of smart city initiatives. This paper addresses the challenge of analyzing and understanding these patterns by employing various clustering techniques on hourly urban traffic flow data. The principal aim is to develop a model that can effectively analyze temporal patterns in urban traffic, uncovering underlying trends and factors influencing traffic flow, which are essential for optimizing smart city infrastructure. To achieve this, we applied DBSCAN, K-Means, Affinity Propagation, Mean Shift, and Gaussian Mixture clustering techniques to the traffic dataset of Aarhus, Denmark's second-largest city. The performance of these clustering methods was evaluated using the Silhouette Score and Dunn Index, with DBSCAN emerging as the most effective algorithm in terms of cluster quality and computational efficiency. The study also compares the training times of the algorithms, revealing that DBSCAN, K-Means, and Gaussian Mixture offer faster training times, while Affinity Propagation and Mean Shift are more computationally intensive. The results demonstrate that DBSCAN not only provides superior clustering performance but also operates efficiently, making it an ideal choice for analyzing urban traffic patterns in large datasets. This research emphasizes the importance of selecting appropriate clustering techniques for effective traffic analysis and management within smart city frameworks, thereby contributing to more efficient urban planning and infrastructure development.

In this paper, the automatic segmentation of multispectral magnetic resonance image of the brain is posed as a clustering problem in the intensity space. Thereafter an automatic clustering technique is proposed to solve this problem. The proposed real-coded variable string length genetic clustering technique (MCVGAPS clustering) is able to evolve the number of clusters present in the data set automatically. Each cluster is divided into several small hyperspherical subclusters and the centers of all these small sub-clusters are encoded in a string to represent the whole clustering. For assigning points to different clusters, these local sub-clusters are considered individually. For the purpose of objective function evaluation, these sub-clusters are merged appropriately to form a variable number of global clusters. A recently developed point symmetry distance based cluster validity index, Sym-index, is optimized to automatically evolve the appropriate number of clusters present in an MR brain image. The proposed method is applied on several simulated T1-weighted, T2- weighted and proton density normal and MS lesion magnetic resonance brain images. Superiority of the proposed method over Fuzzy C-means, Expectation Maximization clustering algorithms are demonstrated quantitatively. The automatic segmentation obtained by multiseed based multiobjective clustering technique (MCVGAPS) is also compared with the available ground truth information.

Medan City is the 3rd most populous city in Indonesia, according to data from the Central Statistics Agency, Medan has a population of 2.49 million in 2022, an increase from the previous 2.46 million in 2021. The increasing number of population inhabiting the city of Medan means that the need for transportation for the people of Medan is also increasing. Trans Metro Deli bus data can be grouped effectively using the mean shift algorithm based on several attributes, namely passenger category, payment method and fare. Each passenger group has different needs and ability to pay, which makes setting fair and efficient fares a challenge. Inappropriate pricing can lead to passenger dissatisfaction, reduce the number of public transportation users, and affect bus operators' revenue. Cluster technique is a well-known clustering technique, which aims to group data into clusters so that each cluster contains data that is as similar as possible. Mean shift belongs to the category of clustering algorithms with unsupervised learning that assigns data points to clusters iteratively by shifting the points towards the mode (mode is the highest density of data points in the region in the context of mean shift). Mean shift does not require determining the number of clusters in advance The attributes used in the clustering process, namely passenger category, payment method and fare can properly create a hyperplane between clusters, thus creating significant differences from each cluster, as evidenced by the silhouette score obtained by 0.64. By conducting this analysis, it is expected to find a more efficient and fair fare clustering pattern, and provide practical recommendations for management in setting fares that are more in line with passenger needs. In addition, this research also aims to evaluate the effectiveness of mean shift clustering in the context of transportation fare analysis.

We propose a split-merge framework for evolutionary clustering. The proposed clustering technique, entitled Split-Merge Evolutionary Clustering is supposed to be more robust to concept drift scenarios by providing the flexibility to consider at each step a portion of the data and derive clusters from it to be used subsequently to update the existing clustering solution. The proposed framework is built around the idea to model two clustering solutions as a bipartite graph, which guides the update of the existing clustering solution by merging some clusters with ones from the newly constructed clustering while others are transformed by splitting their elements among several new clusters. We have evaluated and compared the discussed evolutionary clustering technique with two other state of the art algorithms: a bipartite correlation clustering (PivotBiCluster) and an incremental evolving clustering (Dynamic split-and-merge).

Feature choice involves characteristic a set of the foremost helpful options that produces compatible results because the original entire set of options. A feature choice rule is also evaluated from each the potency and effectiveness points of read. Whereas the potency considerations the time needed to search out a set of options, the effectiveness is expounded to the standard of the set of options. Supported these criteria, an economical Fuzzy C Means (FCM) is projected and by experimentation evaluated in this paper. The quick rule works in 2 steps. Within the commencement, options area unit divided into clusters by exploitation graph-theoretic cluster ways. Within the second step, the foremost representative feature that's powerfully associated with target categories is chosen from every cluster to create a set of options. Options in numerous clusters area unit comparatively freelance; the clustering-based strategy of quick incorporates a high chance of manufacturing a set of helpful and independent options. To make sure the potency of quick, we have a tendency to adopt the economical Fuzzy C Means (FCM) cluster technique. The potency associated effectiveness of the quick rule area unit evaluated through an empirical study. in depth experiments area unit dole out to match quick and a number of other representative feature choice algorithms, namely, FCBF, ReliefF, CFS, Consist, and FOCUS-SF, with relevancy four kinds of well- known classifiers, namely, the chance primarily based Naive Thomas Bayes, the tree-based C4.5, the instance- based IB1,and also the rule-based liquidator before and once feature choice. The results, on thirty five in public on the market real-world high-dimensional image, microarray, and text knowledge, demonstrate that the quick not solely produces smaller subsets of options however additionally improves the performances of the four kinds of classifiers. Index Terms: feature subset selection, relevance, redundancy and high dimensionality.

Fuzzy C-Means (FCM) clustering technique is among the most effective partitional clustering algorithms available in the literature. The Capacitated Vehicle Routing Problem (CVRP) is an important industrial logistics and managerial NP-hard problem. Cluster-First Route-Second Method (CFRS) is one of the efficient techniques used to solve CVRP. In CFRS technique, customers are first divided into clusters in the first phase, then each cluster is solved independently as a Traveling Salesman Problem (TSP) in the second phase. This research is concerned with the clustering phase of CFRS, and TSP is then solved using a traditional optimization method. Three supervised FCM based techniques are proposed to solve the clustering phase at reduced cost via centroids (pre-FCM) initialization phase. The proposed pre-FCM initialization techniques are developed to be problem dependent. Hence, three initialization techniques are first developed using K-means technique, spatially equally distributed, and demand weighted center of mass. Then, a modified demand weighted fuzzy c-means objective function is employed to assign customers to clusters. To compare the performance of the proposed supervised FCM techniques, forty-two CVRP benchmark problems are solved using the traditional fuzzy C-means algorithm and the developed algorithms. Extensive comparisons are conducted between the traditional fuzzy C-means algorithm, the three proposed initialization techniques, and other fuzzy C-means techniques available in the literature. Results show that the proposed three initialization techniques are efficient in terms of solution quality and computational cost.

In recent days, mining data in the form of information and knowledge from large databases is one of the demanding and task. Finding similarity between different attributes in a synthetic dataset is an aggressive concept in data retrieval applications. For this purpose, some of the clustering techniques are proposed in the existing works such as k-means, fuzzy c-means, and fuzzy k-means. But it has some drawbacks that include high overhead, less effective results, computation complexity, high time consumption, and memory utilization. To overcome these drawbacks, a similarity-based categorical data clustering technique is proposed. Here, the similarities of inter- and intra-attributes are simultaneously calculated and it is integrated to improve the performance. The dataset loaded as input, where the preprocessing is performed to remove the noise. Once the data are noise free, the similarity between the elements is computed; then, the most relevant attributes are selected and the insignificant attributes are neglected. The support and confidence measures are estimated by applying association rule mining for resource planning. The similarity-based K-medoids clustering technique is used to cluster the attributes based on the Euclidean distance to reduce the overhead. Finally, the bee colony (BC) optimization technique is used to select the optimal features for further use. In experiments, the results of the proposed clustering system are estimated and analyzed with respect to the clustering accuracy, execution time (s), error rate, convergence time (s), and adjusted Rand index (ARI). From the results, it is observed that the proposed technique provides better results when compared to the other techniques.