Many researchers have begun to study signed networks which are widely existed in real world. In the signed network, the links are labeled the positive or negative sign to represent the active or passive relation betwe...
详细信息
Many researchers have begun to study signed networks which are widely existed in real world. In the signed network, the links are labeled the positive or negative sign to represent the active or passive relation between individuals, such as trusted or distrusted relation in social networks. Communities mining is still a great challenge to the domain of signed networks because of negative links. Unlike communities of unsigned networks, positive links mainly occur in the communities and negative links tend to occur between the communities in the signed networks. Nowadays, many methods which are based on global search for signed network community have been raised, and most of these methods require the global information at each iteration. Besides, determining the number of communities is an important problem for current algorithm for the lack of priori knowledge. To address above problems, a novel community detection method based on local information, is proposed for signed networks in this paper. The proposed method mainly includes two steps. In the first step, the number of communities is determined in terms of the centrality of nodes. In the second step, the local objective function is optimized by the local information of nodes, so the global objective function can also be optimized indirectly. Finally, the communities in signed networks are efficiently found. To validate the proposed method, the comparisons are made with other methods in the synthetic and real signed networks. The experimental results indicate that communities in signed networks can be efficiently found by the proposed method.
Link prediction is essential to both research areas and practical applications. In order to make full use of information of the network, we proposed a new method to predict links in the social network. Firstly, we ext...
Link prediction is essential to both research areas and practical applications. In order to make full use of information of the network, we proposed a new method to predict links in the social network. Firstly, we extracted topological information and attributes of nodes in the social network. Secondly, we integrated them into feature vectors. Finally, we used XGB classifier to predict links using feature vectors. Through expanding information source, experiments on a co-authorship network suggest that our method can improve the accuracy of link prediction significantly.
In order to solve the problem that multi-thresholding segmentation spends too much time finding the optimal solution in medical image segmentation, Otsu multi-thresholding based on dynamic combination of genetic algor...
详细信息
Magnetic resonance imaging(MRI) is a kind of imaging modality, which offers clearer images of soft tissues than computed tomography(CT). It is especially suitable for brain disease detection. It is beneficial to detec...
详细信息
Magnetic resonance imaging(MRI) is a kind of imaging modality, which offers clearer images of soft tissues than computed tomography(CT). It is especially suitable for brain disease detection. It is beneficial to detect diseases automatically and accurately. We proposed a pathological brain detection method based on brain MR images and online sequential extreme learning machine. First, seven wavelet entropies(WE) were extracted from each brain MR image to form the feature vector. Then, an online sequential extreme learning machine(OS-ELM) was trained to differentiate pathological brains from the healthy *** experiment results over 132 brain MRIs showed that the proposed approach achieved a sensitivity of 93.51%, a specificity of 92.22%, and an overall accuracy of 93.33%,which suggested that our method is effective.
Malaria is one of the most serious diseases in the world, which is densely distributed in poverty and remote areas. In the prevention and control of malaria, active surveillance is more efficient than passive surveill...
详细信息
A novel method to measure the graph similarity is proposed, where the labels, in-degrees, and out-degrees of the vertices in the graph are comprehensively considered in order to conquer the high complexity and informa...
详细信息
Locating the source of diffusion is a challenging problem in complex networks and has great practical significance for restraining rumors propagation and controlling epidemics spreading. An efficient locating method s...
详细信息
ISBN:
(纸本)9781509006243
Locating the source of diffusion is a challenging problem in complex networks and has great practical significance for restraining rumors propagation and controlling epidemics spreading. An efficient locating method should have a higher locating accuracy with the minimum required information. Although existing locating methods based on observers consider the time delays of edges, they compute the time delays based on the shortest path, which may differ from the actual diffusion process. Moreover, the higher locating accuracy of traditional method with observers has a great dependence on the assumption that the propagation delays along edges follow a definite distribution such as the Gaussian distribution. In order to solve these shortcomings, this paper proposes a Physarum-inspired method to locate the diffusion source that is independence of the distribution of propagation delays. Our method quantifies the nutrient transportation process in the adaptive network evolved by Physarum, which is used to simulate the information or epidemic diffusion routes in a social network. Simulation results on various benchmark networks show that our method has a better performance in terms of error distance than that of Gaussian method without assuming the definite distribution of time delays. Together with the advantage that our method does not require the sender information of observers compared with existing methods, our method allows for a wider range of applications in the real-world networks.
Gabor filters are generally regarded as the most bionic filters corresponding to the visual perception of human. Their filtered coefficients thus are widely utilized to represent the texture information of irises. How...
详细信息
Gabor filters are generally regarded as the most bionic filters corresponding to the visual perception of human. Their filtered coefficients thus are widely utilized to represent the texture information of irises. However, these wavelet-based iris representations are inevitably being misaligned in iris matching stage. In this paper, we try to improve the characteristics of bionic Gabor representations of each iris via combining the local Gabor features and the key-point descriptors of Scale Invariant Feature Transformation （SIFT）, which respectively simulate the process of visual object class recognition in frequency and spatial domains. A localized approach of Gabor features is used to avoid the blocking effect in the process of image division, meanwhile a SIFT key point selection strategy is provided to remove the noises and probable misaligned key points. For the combination of these iris features, we propose a support vector regression based fusion rule, which may fuse their matching scores to a scalar score to make classification decision. The experiments on three public and self-developed iris datasets validate the discriminative ability of our multiple bionic iris features, and also demonstrate that the fusion system outperforms some state-of-the-art methods.
One of the major problems of axiom pinpointing for incoherent terminologies is the precise positioning within the conflict axioms. In this paper we present a formal notion for the entailment-based axiom pinpointing of...
详细信息
One of the major problems of axiom pinpointing for incoherent terminologies is the precise positioning within the conflict axioms. In this paper we present a formal notion for the entailment-based axiom pinpointing of incoherent terminologies, where the parts of an axiom is defined by atomic entailment. Based on these concepts, we prove the one-to-many relationship between existing axiom pinpointing with the entailment-based axiom pinpointing. For its core task, calculating minimal unsatisfiable entailment, we provide algorithms for OWL DL terminologies using incremental strategy and Hitting Set Tree algorithm. The feasibility of our method is shown by case study and experiment evaluations.
In this paper, a differential evolution (DE) algorithm combined with Lévy flight is proposed to solve the reliability redundancy allocation problems. The Lévy flight is incorporated to enhance the ability of...
详细信息
暂无评论