基于网络嵌入的稀疏子图发现算法

针对稀疏子图发现问题中使用高维稀疏向量表示网络信息存在的时间和空间消耗大的问题，提出一种基于网络嵌入的稀疏子图发现（TGF）算法。该算法首先通过网络嵌入的方法将网络结构映射到低维空间中，得到节点的低维向量表示；然后定义向量空间中的稀疏子集发现问题，将稀疏子图发现问题转化为稀疏子集发现问题；迭代搜索局部密度最低的样本点并对其进行扩张，最终找到一个满足条件的最大稀疏子集。实验结果表明，在Synthetic_1000数据集上与TERA（Triangle and Edge Reduction Algorithm）和WK（Weight of K-hop）算法相比，TGF算法的搜索效率是TERA的1 353倍，是WK算法的4倍，并且在k-line、k-triangle和k-density指标上也取得了较优的结果。     

多视图对比增强的异质图结构学习方法

异质图神经网络作为一种异质图表示学习的方法,可以有效地抽取异质图中的复杂结构与语义信息,在节点分类和连接预测任务上取得了优异的表现,为知识图谱的表示与分析提供了有力的支撑.现有的异质图由于存在一定的噪声交互或缺失部分交互,导致异质图神经网络在节点聚合、更新时融入错误的邻域特征信息,从而影响模型的整体性能.为解决该问题,提出了多视图对比增强的异质图结构学习模型.该模型首先利用元路径保持异质图中的语义信息,并通过计算每条元路径下节点之间特征相似度生成相似度图,将其与元路径图融合,实现对图结构的优化.通过将相似度图与元路径图作为不同视图进行多视图对比,实现无监督信息的情况下优化图结构,摆脱对监督信号的依赖.最后,为解决神经网络模型在训练初期学习能力不足、生成的图结构中往往存在错误交互的问题,设计了一个渐进式的图结构融合方法.通过将元路径图和相似度图递增地加权相加,改变图结构融合过程中相似度图所占的比例,在抑制了因模型学习能力弱引入过多的错误交互的同时,达到了用相似度图中的交互抑制原有干扰交互或补全缺失交互的目的,实现了对异质图结构的优化.选择节点分类与节点聚类作为图结构学习的验证任务,在4种...     

Blockchain Based Consensus Algorithm and Trustworthy Evaluation of Authenticated Subgraph Queries

Over the past era, subgraph mining from a large collection of graph database is a crucial problem. In addition, scalability is another big problem due to insufficient storage. There are several security challenges associated with subgraph mining in today’s on-demand system. To address this downside, our proposed work introduces a Blockchain-based Consensus algorithm for Authenticated query search in the Large-Scale Dynamic Graphs (BCCA-LSDG). The two-fold process is handled in the proposed BCCA-LSDG: graph indexing and authenticated query search (query processing). A blockchain-based reputation system is meant to maintain the trust blockchain and cloud server of the proposed architecture. To resolve the issues and provide safe big data transmission, the proposed technique also combines blockchain with a consensus algorithm architecture. Security of the big data is ensured by dividing the BC network into distinct networks, each with a restricted number of allowed entities, data kept in the cloud gate server, and data analysis in the blockchain. The consensus algorithm is crucial for maintaining the speed, performance and security of the blockchain. Then Dual Similarity based MapReduce helps in mapping and reducing the relevant subgraphs with the use of optimal feature sets. Finally, the graph index refinement process is undertaken to improve the query results. Concerning query error, fuzzy logic is used to refine the index of the graph dynamically. The proposed technique outperforms advanced methodologies in both blockchain and non-blockchain systems, and the combination of blockchain and subgraph provides a secure communication platform, according to the findings.     

An efficient analysis of repetitive structures generated by graph products

In this paper, an efficient method is developed for the analysis of regular structures. A structure is called regular if its model can be formed by a graph product. Here, instead of direct solution of the equations corresponding to a regular structure or finding the inverse of the stiffness matrix directly, modal analysis is used, and eigenvectors are employed for calculating the displacements and then internal forces of the structures. For this purpose, first an efficient method is developed for calculating the eigenvectors of the product graphs, and then a method is presented for using these eigenvectors for evaluating the displacements of a structure. Copyright © 2010 John Wiley & Sons, Ltd.     

基于三角划分的多连通域图形匹配研究

由若干个多边形通过组合产生的具有多个连通域的多连通域图形识别问题是工程图纸识别与理解所要解决的重要问题。首先,通过建立包围多边形、连通多边形概念,完成对多连通域图形的表示;然后,依据最近距离、无遮挡、包含原则添加辅助线,实现三角划分,解决了包含连通的定位问题。在匹配过程中,首先通过匹配包围多边形实现多连通域图形的外轮廓定位;然后通过匹配连通多边形删除连通域,逐步收缩图形,实现图形匹配。应用此方法,解决了多连通域图形的旋转、缩放匹配问题,在汽车线束图纸识别中取得了良好的应用效果。     

Copy-Paste Detection Based on a SIFT Marked Graph Feature Vector

To detect copy-paste tampering, an im-proved SIFT (Scale invariant feature transform)-based al-gorithm was proposed. Maximum angle is defined and a maximum angle-based marked graph is constructed. The m arked graph feature vector is provided to each SIFT key point via discrete polar coordinate transformation. Key points are matched to detect the copy-paste tamper-ing regions. The experimental results show that the pro-posed algorithm can effectively identify and detect the ro-tated or scaled copy-paste regions, and in comparison with the methods reported previously, it is resistant to post-processing, such as blurring, Gaussian white noise and JPEG recompression. The proposed algorithm performs better than the existing algorithm to dealing with scaling transformation.     

网络图自动生成器的设计

本文论述了在网络活动逻辑关系已知的情况下,由计算机绘制网络图时应解决的几个关键性问题。文中首先介绍了一种自动处理网络逻辑关系的原理,其次介绍了生成网络图象的方法,最后给出了一个应用实例。     

A novel proximity graph: Circular neighborhood cell graph for histopathological tissue image analyzing

The cell is the smallest unit of living beings, which has structural and functional properties. Almost all cell behaviors are regulated by various intracellular reactions initiated by the signaling. The signaling and the distance between cells influence each other. Thus, cell-location-based modeling and analyzing of histopathological tissues provide important information to the expert. In literature, methods such as distance-based threshold, K-Nearest Neighbor, Voronoi graphs, Delaunay triangulation, and colored graph have been used. However, circular neighborhood relationships of cells have not been considered by any CAD system so far despite of its crucial impact. Thus, we developed the circular neighborhood cell-graph. Histopathological images of liver were classified by using features extracted from T-Distance, K-Nearest Neighbor, Voronoi, Delaunay, and the proposed cell-graph. Then, the classification performances of the methods were compared. Experimental results show that liver tissue images can be classified with accuracy of 95.7% by using the features provided by the proposed cell-graph model.     

Estimating posterior inference quality of the relational infinite latent feature model for overlapping community detection

Overlapping community detection has become a very hot research topic in recent decades, and a plethora of methods have been proposed. But, a common challenge in many existing overlapping community detection approaches is that the number of communities K must be predefinedmanually. We propose a flexible nonparametric Bayesian generative model for count-value networks, which can allow K to increase as more and more data are encountered instead of to be fixed in advance. The Indian buffet process was used to model the community assignment matrix Z, and an uncollapsed Gibbs sampler has been derived.However, as the community assignment matrix Z is a structured multi-variable parameter, how to summarize the posterior inference results and estimate the inference quality about Z, is still a considerable challenge in the literature. In this paper, a graph convolutional neural network based graph classifier was utilized to help to summarize the results and to estimate the inference quality about Z. We conduct extensive experiments on synthetic data and real data, and find that empirically, the traditional posterior summarization strategy is reliable.     

A multi-objective evolutionary algorithm guided by directed search for dynamic scheduling

This paper considers dynamic multi-objective machine scheduling problems in response to continuous arrival of new jobs, under the assumption that jobs can be rejected and job processing time is controllable. The operational cost and the cost of deviation from the baseline schedule need to be optimized simultaneously. To solve these dynamic scheduling problems, a directed search strategy (DSS) is introduced into the elitist non-dominated sorting genetic algorithm (NSGA-II) to enhance its capability of tracking changing optimums while maintaining fast convergence. The DSS consists of a population re-initialization mechanism (PRM) to be adopted upon the arrival of new jobs and an offspring generation mechanism (OGM) during evolutionary optimization. PRM re-initializes the population by repairing the non-dominated solutions obtained before the disturbances occur, modifying randomly generated solutions according to the structural properties, as well as randomly generating solutions. OGM generates offspring individuals by fine-tuning a few randomly selected individuals in the parent population, employing intermediate crossover in combination with Gaussian mutations to generate offspring, and using intermediate crossover together with a differential evolution based mutation operator. Both PRM and OGM aim to strike a good balance between exploration and exploitation in solving the dynamic multi-objective scheduling problem. Comparative studies are performed on a variety of problem instances of different sizes and with different changing dynamics. Experimental results demonstrate that the proposed DSS is effective in handling the dynamic scheduling problems under investigation.