The aim of this paper is to present interesting results obtained by implementing big data classification algorithm using a cooperative mobile agents model. In this work we focused on the application of this classifica...
详细信息
ISBN:
(纸本)9781479946471
The aim of this paper is to present interesting results obtained by implementing big data classification algorithm using a cooperative mobile agents model. In this work we focused on the application of this classification for Magnetic Resonance Images (MRI) segmentation which is presented based on parallel fine grained c-means algorithm. This algorithm is converged to distributed classification c-means algorithm in order to be implemented on a parallel and distributed virtual machine based on mobile agents. The big data MRI image to be processed is split into small elementary images by the host agent. The obtained list of item images is distributed to mobile agents deployed on each node of the distributed system. Each agent is asked to perform the execution of the c-means algorithm using its assigned image item and return its elementary results to the host agent which computes the current global class centers and newly distributes them. This process is repeated until the convergence of the distributed algorithm. The experimental results show that the distributed classification method can improve the big data segmentation.
Electrocorticogram (EcoG) is a kind of non-stationary signal. Feature extraction and classification are crucial for good EcoG based brain-computer interface (EcoG-BcI). In this paper, a method based on Hilbert-Hung tr...
详细信息
Electrocorticogram (EcoG) is a kind of non-stationary signal. Feature extraction and classification are crucial for good EcoG based brain-computer interface (EcoG-BcI). In this paper, a method based on Hilbert-Hung transform (HHT) is proposed to extract and compress the features of EcoG narrow bands. Then particle swarm optimization (PSO) and G-meansalgorithm are wrapped to adjust feature weights and enhance classification result. Only using 6 channels, the test accuracy of 93% is achieved on Data set I of BcI competition III.
A new method of partitive clustering is developed in the framework of shadowed sets. The core and exclusion regions of the generated shadowed partitions result in a reduction in computations as compared to conventiona...
详细信息
A new method of partitive clustering is developed in the framework of shadowed sets. The core and exclusion regions of the generated shadowed partitions result in a reduction in computations as compared to conventional fuzzy clustering. Unlike rough clustering, here the choice of threshold parameter is fully automated. The number of clusters is optimized in terms of various validity indices. It is observed that shadowed clustering can efficiently handle overlapping among clusters as well as model uncertainty in class boundaries. The algorithm is robust in the presence of outliers. A comparative study is made with related partitive approaches. Experimental results on synthetic as well as real data sets demonstrate the superiority of the proposed approach. (c) 2009 Elsevier Ltd. All rights reserved.
The classification phase is an important step of an automatic fingerprint identification system, where the goal is to restrict only to a subset of the whole database the search time. The proposed system classifies fin...
详细信息
ISBN:
(纸本)9783540748175
The classification phase is an important step of an automatic fingerprint identification system, where the goal is to restrict only to a subset of the whole database the search time. The proposed system classifies fingerprint images in four classes using only directional image information. This approach, unlike the literature approaches, uses the acquired fingerprint image without enhancement phases application. The system extracts only directional image and uses three concurrent decisional modules to classify the fingerprint. The proposed system has a high classification speed and a very low computational cost. The experimental results show a classification rate of 87.27%.
Traditional centralized approaches to security are difficult to apply to multi-agent systems which are used nowadays in e-commerce applications. Developing a notion of trust that is based on the reputation of an agent...
详细信息
ISBN:
(纸本)0819457884
Traditional centralized approaches to security are difficult to apply to multi-agent systems which are used nowadays in e-commerce applications. Developing a notion of trust that is based on the reputation of an agent can provide a softer notion of security that is sufficient for many multi-agent applications. Our paper proposes a mechanism for computing reputation of the trustee agent for use by the trustier agent. The trustier agent computes the reputation based on its own experience as well as the experience the peer agents have with the trustee agents. The trustier agents intentionally interact with the peer agents to get their experience information in the form of recommendations. We have also considered the case of unintentional encounters between the referee agents and the trustee agent, which can be directly between them or indirectly through a set of interacting agents. The clustering is done to filter off the noise in the recommendations in the form of outliers. The trustier agent clusters the recommendations received from referee agents on the basis of the distances between recommendations using the hierarchical agglomerative method. The dendogram hence obtained is cut at the required similarity level which restricts the maximum distance between any two recommendations within a cluster. The cluster with maximum number of elements denotes the views of the majority of recommenders. The center of this cluster represents the reputation of the trustee agent which can be computed using c-means algorithm.
Quality assurance in the powder injection molding is a critical problem due to its complicated processing methods. As surface conditions are major issues for the product quality of the powder injection molding, automa...
详细信息
Quality assurance in the powder injection molding is a critical problem due to its complicated processing methods. As surface conditions are major issues for the product quality of the powder injection molding, automated visual inspection on the surface is highly demanded. This paper proposes representation and recognition schemes for the surface defects on the powder injection molding. From the edge image, line segments were extracted, then they were represented using parameters. Multi-layer perceptron and c-means algorithm were tested to recognize defective features in the powder injection molding. The neural network method showed better recognition for the defective features based on the selected measures. Significance: The surface defect in powder injection molding is a critical problem for the product quality assurance. From the complicated surface features, the recognition of defective features were compared between an artificial neural network and traditional pattern recognition method.
In this paper, a new non-iterative clustering method is proposed. It consists of two passes. In the first pass, the mean distance from one object to its nearest neighbor is estimated. Based on this distance, those noi...
详细信息
In this paper, a new non-iterative clustering method is proposed. It consists of two passes. In the first pass, the mean distance from one object to its nearest neighbor is estimated. Based on this distance, those noises far away from objects are extracted and removed. In the second pass, the mean distance from the remaining objects to their nearest neighbors is computed. Based on the distance, all the intrinsicclusters are then found. The proposed method is non-iterative and can automatically determine the number of clusters. Experimental results also show that the partition generated by the proposed method is more reasonable than that of the well-known c-means algorithm in many complicated object distributions.
暂无评论