Search

Neutron clustering is a recently identified problem with Monte Carlo eigenvalue calculations which can produce significantly erroneous results. Previous work by Sutton & Mittal (2017) considered neutron clustering as a problem of maintaining 'genetic diversity' within the neutron population. This paper proposes reducing the extent of neutron clustering by replacing fission neutrons in the source bank with uncorrelated neutrons, sampled from a uniform source distribution – effectively adding new neutron genealogies to the population. The efficacy of the method is demonstrated on a number of simple problems, showing improved behaviour of the Shannon entropy and neutron centre-of-mass. Although currently limited in scope, this paper intends to provide a route to reducing clustering effects in more general problems. [ABSTRACT FROM AUTHOR]

Multi-source contamination mapping is a critical aspect of radiation detection and environmental monitoring. This paper introduces an innovative algorithm for accurate and efficient multi-source contamination mapping. The algorithm comprises two main components: (1) search area identification and (2) source localization using a particle filter. The search area identification involves extrapolating directional measurements to create search areas, employing clustering algorithms to refine these areas, and generating final search areas based on weighted factors. The incorporation of directional measurements significantly enhances performance by reducing parameter search space of the particle filter and thus improving practical application potential. The particle filter-based source localization employs a four-stage process, involving generation of initial guesses, comparison of expected and acquired measurements, employing genetic algorithms for particle selection, and a convergence criterion. The algorithm's efficacy is demonstrated through four simulated experimental scenarios, each with varying numbers and densities of point radioactive sources. Results illustrate successful search area identification and source localization with an average error of approximately 1 meter and source activity estimation of within 25% of the ground-truth values. Further improvements to the area clustering algorithm, improved mission planning for online applications, and extensive testing of the algorithm are planned in the future. With these improvements, the algorithm presented here could enhance the ability to respond to radioactive contamination incidents swiftly and effectively, ensure timely mapping of contaminated areas, and maintain human safety in the event of an incident. [ABSTRACT FROM AUTHOR]

Cluster analysis is an important exploratory data analysis technique used in many different fields, such as data mining, pattern recognition, grouping, decision science, machine learning, psychology and statistics. K-means algorithm is a popular, simple and efficient clustering algorithm. However, the algorithm may suffer from several drawbacks, such as converging to local optima and depending on the assignment of initial centroids. Metaheuristic algorithms are frequently used for cluster analysis to overcome the aforementioned drawbacks of conventional clustering algorithms. Glowworm Swarm Optimization (GSO) algorithm is a swarm intelligence based optimization algorithm to find multiple global solutions. In this paper, an approach for cluster analysis based on K-means and GSO has been presented. The proposed approach is compared to K-means, Fuzzy c-means, GSO algorithms with well-known benchmark datasets, such as Iris, Breast Cancer, Ecoli, Diabetes, Haberman's Survival Data in terms of cluster validation measures, such as F-measure and rand index. [ABSTRACT FROM AUTHOR]

Email is the most common and comparatively the most efficient means of exchanging information in today's world. However, given the widespread use of emails in all sectors, they have been the target of spammers since the beginning. Filtering spam emails has now led to critical actions such as forensic activities based on mining spam email. The data mine for spam emails at the University of Alabama at Birmingham is considered to be one of the most prominent resources for mining and identifying spam sources. It is a widely researched repository used by researchers from different global organizations. The usual process of mining the spam data involves going through every email in the data mine and clustering them based on their different attributes. However, given the size of the data mine, it takes an exceptionally long time to execute the clustering mechanism each time. In this paper, we have illustrated sampling as an efficient tool for data reduction, while preserving the information within the clusters, which would thus allow the spam forensic experts to quickly and effectively identify the 'hot zone' from the spam campaigns. We have provided detailed comparative analysis of the quality of the clusters after sampling, the overall distribution of clusters on the spam data, and timing measurements for our sampling approach. Additionally, we present different strategies which allowed us to optimize the sampling process using data-preprocessing and using the database engine's computational resources, and thus improving the performance of the clustering process. [ABSTRACT FROM AUTHOR]

We present a new algorithm for locating a small cluster of points with differential privacy [Dwork, McSherry, Nissim, and Smith, 2006]. Our algorithm has implications to private data exploration, clustering, and removal of outliers. Furthermore, we use it to significantly relax the requirements of the sample and aggregate technique [Nissim, Raskhodnikova, and Smith, 2007], which allows compiling of “off the shelf” (non-private) analyses into analyses that preserve differential privacy., Engineering and Applied Sciences

Sensing, processing and communication abilities of sensor nodes, make them applicable for many sensor network applications. Because of the energy constraints, several clustering approaches are proposed to decrease energy consumption. In this study, we firstly present the importance of the clustering technique for sensor networks and describe the genetic algorithm. And then, we describe most up to date genetic algorithm based clustering approaches for wireless sensor networks; providing a detailed comparison table and highlighting their objectives and characteristics. [ABSTRACT FROM AUTHOR]