首页期刊导航|International journal of software science and computational intelligence
期刊信息/Journal information
International journal of software science and computational intelligence
IGI Publishing
IGI Publishing
季刊
1942-9045
International journal of software science and computational intelligence/Journal International journal of software science and computational intelligence
查看更多>>摘要：One of the main problems in K-means clustering is setting of initial centroids which can cause misclustering of patterns which affects clustering accuracy. Recently, a density and distance-based technique for determining initial centroids has claimed a faster convergence of clusters. Motivated from this key idea, the authors study the impact of initial centroids on clustering accuracy for unsupervised feature selection. Three metrics are used to rank the features of a data set. The centroids of the clusters in the data sets, to be applied in K-means clustering, are initialized randomly as well as by density and distance-based approaches. Extensive experiments are performed on 15 datasets. The main significance of the paper is that the K-means clustering yields higher accuracies in majority of these datasets using proposed density and distance-based approach. As an impact of the paper, with fewer features, a good clustering accuracy can be achieved which can be useful in data mining of data sets with thousands of features.
查看更多>>摘要：There are many similarities in the symptoms of several types of cancer and that makes it sometimes difficult for the physicians to do an accurate diagnosis. In addition, it is a technical challenge to classify accurately the cancer cells in order to differentiate one type of cancer from another. The DNA microarray technique (also called the DNA chip) has been used in the past for the classification of cancer but it generates a large volume of noisy data that has many features, and is difficult to analyze directly. This paper proposes a new method, combining the genetic algorithm, case-based reasoning, and the k-nearest neighbor classifier, which improves the performance of the classification considerably. The authors have also used the well-known Mahalanobis distance of multivariate statistics as a similarity measure that improves the accuracy. A case-based classifier approach together with the genetic algorithm has never been applied before for the classification of cancer, same with the application of the Mahalanobis distance. Thus, the proposed approach is a novel method for the cancer classification. Furthermore, the results from the proposed method show considerably better performance than other algorithms. Experiments were done on several benchmark datasets such as the leukemia dataset, the lymphoma dataset, ovarian cancer dataset, and breast cancer dataset.
查看更多>>摘要：Load balancing is the phenomenon of distributing workload over various computing resources efficiently. It offers enterprises to efficiently manage different application or workload demands by allocating available resources among different servers, computers, and networks. These services can be accessed and utilized either for home use or for business purposes. Due to the excessive load on the cloud, sometimes it is not feasible to offer all these services to different users efficiently. To solve this excessive load issue, an efficient load balancing technique is used to offer satisfactory services to users as per their expectations also leading to efficient utilization of resources and applications on the cloud platform. This paper presents an enhanced load balancing algorithm named as a two-phase load balancing algorithm. It uses a two-phase checking load balancing approach where the first phase is to divide all virtual machines into two different tables based on their state, that is, available or busy while in the second phase, it equally distributes the loads. The various parameters used to measure the performance of the proposed algorithm are cost, data center processing time, and response time. Cloud analyst simulation tool is used to simulate the algorithm. Simulation results demonstrate superiority of the algorithm with existing ones.
查看更多>>摘要：In the world of the internet of things (IoT), many connected objects generate an enormous amount of data. This data is used to analyze and make decisions about specific phenomena. If an object generates wrong data, it will influence the analysis of this collected data and the decision later. A forensics analysis is necessary to detect IoT nodes that are failing. This paper deals with a problem: the detection of these nodes, which generate erroneous data. The study starts to collect in a cloud computing server temperature measurements (the case study); using temperature sensors, the communication of the nodes is based on the HIP (host identity protocol). The detection is made using a data mining classification technique, in order to group the connected objects according to the collected measurements. At the end of the study, very good results were found, which opens the door to further studies.
查看更多>>摘要：Deep learning (DL) has been used globally in almost every sector of technology and society. Despite its huge success, DL models and applications have been susceptible to adversarial attacks, impacting the accuracy and integrity of these models. Many state-of-the-art models are vulnerable to attacks by well-crafted adversarial examples, which are perturbed versions of clean data with a small amount of noise added, imperceptible to the human eyes, and can quite easily fool the targeted model. This paper introduces six most effective gradient-based adversarial attacks on the ResNet image recognition model, and demonstrates the limitations of traditional adversarial retraining technique. The authors then present a novel ensemble defense strategy based on adversarial retraining technique. The proposed method is capable of withstanding the six adversarial attacks on cifar10 dataset with accuracy greater than 89.31% and as high as 96.24%. The authors believe the design methodologies and experiments demonstrated are widely applicable to other domains of machine learning, DL, and computation intelligence securities.
查看更多>>摘要：In this article, the author investigates information processing algorithms in order to determine the relative UAV motion parameters in a group flight and proposes an algorithm for estimating the leading UAV motion parameters from the results of relative motion measurements. Such researches are especially important nowadays in all the spheres where the drones are used and/or will be used. The author considers the problems of management of the UAV in the group flight, formulation of the problem for processing of the information in such the conditions. The article considers relative motion equation and the synthesis of information processing algorithms in the master-slave model of the flying group of UAV.
NETL
NSTL
Igi Global