Hybrid Of K-means Research Articles

Hybrid of K-Means and partitioning around medoids for predicting COVID-19 cases: Iraq case study

COVID-19 was discovered near the end of 2019 in Wuhan, China. In a short period, the virus had spread throughout the entire world. One of the primary concerns of managers and decision-makers in all types of hospitals nowadays is to implement detection plans for status of patient (Negative, Positive) in order to provide enough care at the proper moment. To reduce a pandemic of COVID-19, improving health care quality could be advantageous. Making clusters of patients with similar features and symptoms supplies an overview of health quality given to similar patients. In the scope of medical machine learning, the K-means and Partitioning Around Medoids (PAM) clustering algorithms are usually used to produce clusters depend on similarity and to detect helpful patterns from sizes of data. In this paper, we proposed a hybrid algorithm of K-Means and Partitioning Around Medoids (PAM) called K-MP to take benefits of both PAM and K-Means to construct an efficient model for predicting patient status. The suggested model for the real dataset was collected from 400 patients in the many Iraqi clinics using a questionnaire. We evaluated the proposed K-MP by using true negative rate, balance accuracy, precision, accuracy, recall, mean absolute error, F1 score, and root mean square error. From these performance measures, we found that K-MP is more efficient in discovering patient status comparing to K-Means and PAM.

A hybrid of K-means and genetic algorithm to solve a bi-objective green delivery and pick-up problem

ABSTRACT Emissions of hazardous greenhouse gases from vehicles poses a remarkable threat to the environment. This study considers a bi-objective green delivery and pick-up problem wherein vehicle fuel burnt per distance, as a CO2 emission metric, and fixed costs of the fleet are minimized. A mathematical model is devised to obtain exact solutions. A hybrid three-step metaheuristic approach is devised to tackle large-size instances. To generate initial solutions, customers are clustered based on their locations using k-means algorithm. Afterward, a genetic algorithm is used for solving a traveling salesman problem within each cluster. Finally, NSGA-II is incorporated to concatenate clusters, obtained from the initial solution, while generating non-dominated solutions by performing a trade-off between costs and emissions. Random problem instances are generated and solved to make a comparison between the performance of hybrid methodology against NSGA-II, MOPSO, and multi-objective dragonfly algorithm. Results indicate the hybrid approach’s superiority to others.

A constraint programming approach and a hybrid of genetic and K-means algorithms to solve the p-hub location-allocation problems

ABSTRACT p-Hub location-allocation problem is one of the most interesting subjects in the location theory. Hubs act as switching points to reduce the transportation cost. In this study, two new solution methods, a constraint programming (CP) based model and a hybrid of k-means and genetic algorithm (KGA), are developed to generate exact and approximate solutions, respectively. The proposed CP formulation is more understandable and straightforward in comparison with the MIP model. The experimental results indicate that the CP model uses the memory of the computer (RAM) more efficiently, which enables us to solve the medium size problems. But, in terms of run time, this method cannot be superior to the MIP model. The CP formulation is also extended for the multi allocation p-hub location problem. K-means algorithm, a well-known algorithm for clustering data, is used to generate initial solutions of GA. Furthermore, a new adaptive crossover operator, which is based on the k-means algorithm, is proposed. The experimental results indicate that the KGA algorithm is superior to the GA, regarding time, objective value, and quality of solution measures.

Enhance Clustering Algorithm Using Optimization

Unsupervised learning can reveal the structure of datasets without being concerned with any labels, K-means clustering is one such method. Traditionally the initial clusters have been selected randomly, with the idea that the algorithm will generate better clusters. However, studies have shown there are methods to improve this initial clustering as well as the K-means process. This paper examines these results on different types of datasets to study if these results hold for all types of data. Another method that is used for unsupervised clustering is the algorithm based on Particle Swarm Optimization. For the second part this paper studies the classic K-means based algorithm and a Hybrid K-means algorithm which uses PSO to improve the results from K-means. The hybrid K-means algorithms are compared to the standard K-means clustering on two benchmark classification problems. In this project we used Kaggle dataset to with different size (small, large and medium) for comparison PSO, k-means and k-means hybrid.

Prediction of Breast Cancer from Imbalance Respect Using Cluster-Based Undersampling Method.

To overcome the two-class imbalanced problem existing in the diagnosis of breast cancer, a hybrid of K-means and Boosted C5.0 (K-Boosted C5.0) is proposed which is based on undersampling. K-means is utilized to select the informative samples near the boundary. During the training phase, the K-means algorithm clusters the majority and minority instances and selects a similar number of instances from each cluster. Boosted C5.0 is then used as the classifier. As there is one different instance selection factor via clustering that encourages the diversity of the training subspace in K-Boosted C5.0, it would be a great advantage to get better performance. To test the performance of the new hybrid classifier, it is implemented on 12 small-scale and 2 large-scale datasets, which are the often used datasets in class imbalanced learning. The extensive experimental results show that our proposed hybrid method outperforms most of the competitive algorithms in terms of Matthews' correlation coefficient (MCC) and accuracy indices. It can be a good alternative to the well-known machine learning methods.

Optimal cluster analysis using hybrid K-Means and Ant Lion Optimizer

K-Means is a popular cluster analysis method which aims to partition a number of data points into K clusters. It has been successfully applied to a number of problems. However, the efficiency of K-Means depends on its initialization of cluster centers. Different swarm intelligence techniques are applied to clustering problem for enhancing the performance. In this work a hybrid clustering approach based on K-means and Ant Lion Optimization has been considered for optimal cluster analysis. Ant Lion Optimization (ALO) is a stochastic global optimization model. The performance of the proposed algorithm is compared against the performance of K-Means, KMeans-PSO, KMeans-FA, DBSCAN and Revised DBSCAN clustering methods based on different performance metrics. Experimentation is performed on eight datasets, for which the statistical analysis is carried out. The obtained results indicate that the hybrid of K-Means and Ant Lion Optimization method performs preferably better than the other three algorithms in terms of sum of intracluster distances and F-measure.

Hybridization of K-Means and Firefly Algorithm for intrusion detection system

During the last decade, anomaly detection has attracted the attention of many researchers to overcome the weakness of signature-based IDSs in detecting novel attacks. Indeed, it is difficult to provide secure information systems and to maintain them in a secure state during their lifetime. An IDS is a device or software application that monitors network or system activities for malicious task or policy violations and produces reports to a management station. A metaheuristic is a high-level problem independent algorithmic framework. These are problem-independent techniques and do not take advantage of any specificity of the problem. The main aim of meta-heuristic algorithms is to quickly find solution to a problem. This solution may not be the best of all possible solutions to the problem but still they stand valid as they do not require excessively long time to be solved. Firefly Algorithm is one of the new metaheuristic algorithms for optimization problems inspired by the flashing behavior of fireflies. In this work, a new algorithm for anomaly detection has been introduced which is a hybridization of K-Means and Firefly Algorithm. The algorithm uses clustering to build the training model and uses classification to evaluate on the test set. The subject algorithm is evaluated on the NSL-KDD dataset, which is quite impressive. Further, a comparison study has been performed between the newly developed algorithm with other clustering algorithms including K-Means + Cuckoo, K-Means + Bat, K-Means, K-Means++, Canopy and Farthest First. The results show that K-Means + Firefly and K-Means + Bat outperforms by a huge margin.

Data clustering using seed disperser ant algorithm

Nature-inspired optimization algorithms have become popular in the past decade. They have been applied to solve various kinds of problems. Among these would be data clustering, which has become popular in data mining in recent times due to the data explosion. In the last decade, many metaheuristic algorithms have been used to obtain improved data clustering optimization for solving data mining problems. In this paper, we applied the seed disperser ant algorithm (SDAA), which mimics the evolution of an Aphaenogaster senilis ant colony, and we introduced a modified SDAA that is a hybrid of K-means and SDAA for solving data clustering problems. The solutions obtained for the data clustering are very promising in terms of quality of solutions and convergence speed of the algorithm.

Design of an Expert System for Distribution Planning System using Soft Computing Techniques

An expert system for distribution planning is proposed in this paper. Choosing a best location of a distribution substation and grouping the various load points to be fed from a particular distribution substation has always been a concern to the distribution planners. Here in this paper the authors present a hybridization of K-means clustering method with fuzzy context aware decision algorithm for choosing the optimum location of distribution substation and its feeder layout. K means clustering has been applied to various loads which are at different location to form a cluster with load points in closer proximity so that a substation could be placed for each cluster for the distribution of power. Fuzzy Context Aware Decision Algorithm based on the Analytical Hierarchy process (AHP) is then applied on each cluster to decide on the feeder layout connecting the load points in each cluster. The feeder layout is based on the various reliability factors and thus the result obtained will lead to optimum feeder path and will hence lower long range distribution expenses.

A Model for Predicting Students’ Academic Performance using a Hybrid of K-means and Decision tree Algorithms

A Model for Predicting Students’ Academic Performance using a Hybrid of K-means and Decision tree Algorithms

Ranking banks using K-Means and Grey relational method

Article history: Received January 20, 2014 Accepted 30 August 2014 Available online September 1 2014 Research has been done on evaluating the performance more emphasis either on the technical aspects of the performance evaluation or considers only one aspect of banks such as sales, revenue, etc. This paper presents an empirical investigation to measure the relative efficiency of different banks using a hybrid of K-means and Grey relational analysis. The study uses eight criteria such as resources, customer satisfaction, etc. for clustering different banks and using Grey relational analysis provides ranking of different branches of a bank named Bank Qarzol Hasane Resalat located in city of Tehran, Iran. The preliminary study has indicated that the proposed model of this paper may capable of providing some promising results. © 2014 Growing Science Ltd. All rights reserved. Clustering K-Means Grey relational Banks Ranking

Breast cancer diagnosis based on feature extraction using a hybrid of K-means and support vector machine algorithms

With the development of clinical technologies, different tumor features have been collected for breast cancer diagnosis. Filtering all the pertinent feature information to support the clinical disease diagnosis is a challenging and time consuming task. The objective of this research is to diagnose breast cancer based on the extracted tumor features. Feature extraction and selection are critical to the quality of classifiers founded through data mining methods. To extract useful information and diagnose the tumor, a hybrid of K-means and support vector machine (K-SVM) algorithms is developed. The K-means algorithm is utilized to recognize the hidden patterns of the benign and malignant tumors separately. The membership of each tumor to these patterns is calculated and treated as a new feature in the training model. Then, a support vector machine (SVM) is used to obtain the new classifier to differentiate the incoming tumors. Based on 10-fold cross validation, the proposed methodology improves the accuracy to 97.38%, when tested on the Wisconsin Diagnostic Breast Cancer (WDBC) data set from the University of California – Irvine machine learning repository. Six abstract tumor features are extracted from the 32 original features for the training phase. The results not only illustrate the capability of the proposed approach on breast cancer diagnosis, but also shows time savings during the training phase. Physicians can also benefit from the mined abstract tumor features by better understanding the properties of different types of tumors.