Pathfinder: Visual Analysis of Paths in Graphs

The analysis of paths in graphs is highly relevant in many domains. Typically, path‐related tasks are performed in node‐link layouts. Unfortunately, graph layouts often do not scale to the size of many real world networks. Also, many networks are multivariate, i.e., contain rich attribute sets associated with the nodes and edges. These attributes are often critical in judging paths, but directly visualizing attributes in a graph layout exacerbates the scalability problem. In this paper, we present visual analysis solutions dedicated to path‐related tasks in large and highly multivariate graphs. We show that by focusing on paths, we can address the scalability problem of multivariate graph visualization, equipping analysts with a powerful tool to explore large graphs. We introduce Pathfinder, a technique that provides visual methods to query paths, while considering various constraints. The resulting set of paths is visualized in both a ranked list and as a node‐link diagram. For the paths in the list, we display rich attribute data associated with nodes and edges, and the node‐link diagram provides topological context. The paths can be ranked based on topological properties, such as path length or average node degree, and scores derived from attribute data. Pathfinder is designed to scale to graphs with tens of thousands of nodes and edges by employing strategies such as incremental query results. We demonstrate Pathfinder's fitness for use in scenarios with data from a coauthor network and biological pathways.     

GlyphLink: An interactive visualization approach for semantic graphs

Graph analysis by data visualization involves achieving a series of topology-based tasks. When the graph data belongs to a data domain that contains multiple node and link types, as in the case of semantic graphs, topology-based tasks become more challenging. To reduce visual complexity in semantic graphs, we propose an approach which is based on applying relational operations such as selecting and joining nodes of different types. We use node aggregation to reflect the relational operations to the graph. We introduce glyphs for representing aggregated nodes. Using glyphs lets us encode connectivity information of multiple nodes with a single glyph. We also use visual parameters of the glyph to encode node attributes or type specific information. Rather than doing the operations in the data abstraction layer and presenting the user with the resulting visualization, we propose an interactive approach where the user can iteratively apply the relational operations directly on the visualization. We present the efficiency of our method by the results of a usability study that includes a case study on a subset of the International Movie Database. The results of the controlled experiment in our usability study indicate a statistically significant contribution in reducing the completion time of the evaluation tasks.     

语义增强的大规模多元图简化可视分析方法

网络图可视化可以有效展示网络节点之间的连接关系,广泛应用于诸多领域,如社交网络、知识图谱、生物基因网络等.随着网络数据规模的不断增加,如何简化表达大规模网络图结构已成为图可视化领域中的研究热点.经典的网络图简化可视化方法主要包括图采样、边绑定和图聚类等技术,在减少大量点线交叉造成的视觉紊乱的基础上,提高用户对大规模网络结构的探索和认知效率.然而,上述方法主要侧重于网络图中的拓扑结构,却较少考虑和利用多元图节点的多维属性特征,难以有效提取和表达语义信息,从而无法帮助用户理解大规模多元网络的拓扑结构与多维属性之间的内在关联,为大规模多元图的认知和理解带来困难.因此,本文提出一种语义增强的大规模多元图简化可视分析方法,首先在基于模块度的图聚类算法基础上提取出网络图的层次结构;其次通过多维属性信息熵的计算和比较分析,对网络层次结构进行自适应划分,筛选出具有最优属性聚集特征的社团;进而设计交互便捷的多个关联视图来展示社团之间的拓扑结构、层次关系和属性分布,从不同角度帮助用户分析多维属性在社团形成和网络演化中的作用.大量实验结果表明,本文方法能够有效简化大规模多元图的视觉表达,可以快速分析不同应用领域大规模多元图的关联结构与语义构成,具有较强的实用性.     

基于时尚图谱增强的个性化互补服装推荐

近年来,人们对时尚穿搭有了更高的美学追求。因此,个性化互补服装推荐,即为用户推荐与他/她已购买服装相匹配的互补服装,逐渐吸引了学术界的广泛关注。个性化互补服装推荐不同于一般的推荐任务（如电影推荐）,它推荐的服装需要满足两个条件：1）与目标服装搭配;2）满足用户偏好。因此,相关的现有方法主要是基于时尚单品的多模态数据,着力于建模单品与单品之间的兼容性交互和用户与单品之间的偏好交互,以实现个性化互补服装的推荐。这些方法的缺点主要在于它们将每一个单品-单品交互或者用户-单品交互看作一个独立的数据实例,而忽略了单品的属性知识以及时尚实体（即,用户、单品及属性）之间的高阶交互关系。事实上,与一个单品（如,上衣）搭配的所有互补单品（如,下衣）可能会共享某些相同的属性（如,颜色）;同时,具有类似品味的用户也可能倾向于选择具有类似属性特征的单品。显然,这些时尚实体之间的高阶关系蕴含了丰富的有关单品兼容性和用户偏好的协同信号,因而能够促进个性化互补服装推荐模型性能的提升。据此,本文构建了一个大规模协同时尚图谱,并基于图卷积神经网络（Graph Neural Networks,GNNs）来探索时尚实体之间的高阶关系,进而更好地实现个性化的互补服装推荐。具体地,本文提出了一个新颖的基于时尚图谱增强的个性化互补服装推荐模型（Fashion Graph-enhanced Personalized Complementary Clothing Recommendation）,简称为FG-PCCR。FG-PCCR由两个关键的部分组成：独立的一阶交互建模和协同的高阶交互建模。一方面,独立的一阶交互建模模块基于视觉和文本模态数据,致力于通过神经网络和矩阵分解方法分别对单品-单品搭配交互和用户-单品偏好交互进行综合性建模。另一方面,协同的高阶交互模块基于构建的协同的时尚图谱,通过图神经网络利用信息传播机制来提取高阶的协同信号,进一步丰富用户和单品的向量表示。FG-PCCR模型能够有效整合时尚实体之间的复杂的高阶关系信息,用户和单品的表示学习,进而改进个性化互补推荐的效果。最后,对于给定的用户和目标上衣,我们能够得到推荐的下衣的个性化兼容性分数。另外,在真实的时尚数据集上做的大量实验,充分地验证了本文所提模型FG-PCCR相对于基准方法的优越性。     

基于最小描述长度原则的属性图概要方法

图概要技术是管理、分析和可视化大规模图的关键技术之一。如何综合结构和属性信息进行图概要是一个挑战。大部分现有的图概要方法或者只考虑结构或属性某一方面的信息,或者要求属性的表现形式是一致的。结合信息论中最小描述长度原则,对属性图概要问题建模,将其转化为求解最小表示代价问题,以实现图压缩和图概要的双重目标。提出了一种计算节点属性相似性的方法,该属性度量方法对节点属性的限制较小,并且将节点间的相似性统一为存储代价,实现了节点结构相似和属性相似的协同考虑。提出了两种求解最小代价表示的图概要算法。在真实和合成的数据集上实验,验证了提出算法的有效性。     

The Readability of Path‐Preserving Clusterings of Graphs

Graph visualization systems often exploit opaque metanodes to reduce visual clutter and improve the readability of large graphs. This filtering can be done in a path‐preserving way based on attribute values associated with the nodes of the graph. Despite extensive use of these representations, as far as we know, no formal experimentation exists to evaluate if they improve the readability of graphs. In this paper, we present the results of a user study that formally evaluates how such representations affect the readability of graphs. We also explore the effect of graph size and connectivity in terms of this primary research question. Overall, for our tasks, we did not find a significant difference when this clustering is used. However, if the graph is highly connected, these clusterings can improve performance. Also, if the graph is large enough and can be simplified into a few metanodes, benefits in performance on global tasks are realized. Under these same conditions, however, performance of local attribute tasks may be reduced.     

Pre-Train and Learn:Preserving Global Information for Graph Neural Networks

Graph neural networks(GNNs) have shown great power in learning on graphs.However,it is still a challenge for GNNs to model information faraway from the source node.The ability to preserve global information can enhance graph representation and hence improve classification precision.In the paper,we propose a new learning framework named G-GNN(Global information for GNN) to address the challenge.First,the global structure and global attribute features of each node are obtained via unsupervised pre-training,and those global features preserve the global information associated with the node.Then,using the pre-trained global features and the raw attributes of the graph,a set of parallel kernel GNNs is used to learn different aspects from these heterogeneous features.Any general GNN can be used as a kernal and easily obtain the ability of preserving global information,without having to alter their own algorithms.Extensive experiments have shown that state-of-the-art models,e.g.,GCN,GAT,Graphsage and APPNP,can achieve improvement with G-GNN on three standard evaluation datasets.Specially,we establish new benchmark precision records on Cora(84.31％) and Pubmed(80.95％) when learning on attributed graphs.     

基于双向对齐与属性信息的跨语言实体对齐

实体对齐表示在不同的知识图谱中查找引用相同现实身份的实体。目前主流的基于图嵌入的实体对齐方法中的对齐实体通常具有相似的属性,有效利用属性信息可提升实体对齐效果,同时由于不同知识图谱之间的知识分布差异,仅考虑单个方向的对齐预测会导致预测结果出现偏差。针对上述问题,提出一种改进的跨语言实体对齐方法。利用融合属性信息的双向对齐图卷积网络模型,将前馈神经网络编码实体对应的属性信息与初始的实体嵌入相结合,得到联合属性信息的实体表示,并使用双向对齐机制实现跨语言的实体对齐预测。在3个跨语言数据集上的实验结果表明,该方法通过融合更多的知识图谱信息增强了实体表示能力,并且利用双向对齐机制缓解了数据分布差异问题,相比基于图嵌入的实体对齐方法整体性能更优。     

WYTIWYR: A User Intent-Aware Framework with Multi-modal Inputs for Visualization Retrieval

Retrieving charts from a large corpus is a fundamental task that can benefit numerous applications such as visualization recommendations. The retrieved results are expected to conform to both explicit visual attributes (e.g., chart type, colormap) and implicit user intents (e.g., design style, context information) that vary upon application scenarios. However, existing example-based chart retrieval methods are built upon non-decoupled and low-level visual features that are hard to interpret, while definition-based ones are constrained to pre-defined attributes that are hard to extend. In this work, we propose a new framework, namely WYTIWYR (What-You-Think-Is-What-You-Retrieve), that integrates user intents into the chart retrieval process. The framework consists of two stages: first, the Annotation stage disentangles the visual attributes within the query chart; and second, the Retrieval stage embeds the user's intent with customized text prompt as well as bitmap query chart, to recall targeted retrieval result. We develop aprototype WYTIWYR system leveraging a contrastive language-image pre-training (CLIP) model to achieve zero-shot classification as well as multi-modal input encoding, and test the prototype on a large corpus with charts crawled from the Internet. Quantitative experiments, case studies, and qualitative interviews are conducted. The results demonstrate the usability and effectiveness of our proposed framework.     

基于因子图模型的动态图半监督聚类算法

针对动态图的聚类主要存在着两点不足:首先, 现有的经典聚类算法大多从静态图分析的角度出发, 无法对真实网络图持续演化的特性进行有效建模, 亟待对动态图的聚类算法展开研究, 通过对不同时刻图快照的聚类结构进行分析进而掌握图的动态演化情况.其次, 真实网络中可以预先获取图中部分节点的聚类标签, 如何将这些先验信息融入到动态图的聚类结构划分中, 从而向图中的未标记节点分配聚类标签也是本文需要解决的问题.为此, 本文提出进化因子图模型(Evolution factor graph model, EFGM)用于解决动态图节点的半监督聚类问题, 所提EFGM不仅可以捕获动态图的节点属性和边邻接属性, 还可以捕获节点的时间快照信息.本文对真实数据集进行实验验证, 实验结果表明EFGM算法将动态图与先验信息融合到一个统一的进化因子图框架中, 既使得聚类结果满足先验知识, 又契合动态图的整体演化规律, 有效验证了本文方法的有效性.     

Combine and Conquer

We present a general technique for dynamizing a class of problems whose underlying structure is a computation graph embedded in a tree. We introduce three fully dynamic data structures, called path attribute systems, tree attribute systems, and linear attribute grammars, which extend and generalize the dynamic trees of Sleator and Tarjan. More specifically, we associate values, called attributes, with the nodes and paths of a rooted tree. Path attributes form a path attribute system if they can be maintained in constant time under path concatenation. Node attributes form a tree attribute system if the tree attributes of the tail of a path Π can be determined in constant time from the path attributes of Π. A linear attribute grammar is a tree-based linear expression such that the values of a node μ are calculated from the values at the parent, siblings, and/for children of μ. We provide a framework for maintaining path attribute systems, tree attribute systems, and linear attribute grammars in a fully dynamic environment using linear space and logarithmic time per operation. Also, we demonstrate the applicability of our techniques by showing examples of graph and geometric problems that can be efficiently dynamized, including biconnectivity and triconnectivity queries, planarity testing, drawing trees and series-parallel digraphs, slicing floorplan compaction, point location, and many optimization problems on bounded tree-width graphs. Received May 13, 1994; revised October 12, 1995.     

基于节点-属性二部图的网络表示学习模型

在图结构数据上开展推理计算是一项重大的任务,该任务的主要挑战是如何表示图结构知识使机器可以快速理解并利用图数据。对比现有表示学习模型发现,基于随机游走方法的表示学习模型容易忽略属性对节点关联关系的特殊作用,因此提出一种基于节点邻接关系与属性关联关系的混合随机游走方法。首先通过邻接节点间的共同属性分布计算属性权重,并获取节点到每个属性的采样概率;然后分别从邻接节点与含有共有属性的非邻接节点中提取网络信息;最后构建基于节点-属性二部图的网络表示学习模型,并通过上述采样序列学习得到节点向量表达。在Flickr、BlogCatalog、Cora公开数据集上,用所提模型得到的节点向量表达进行节点分类的Micro-F1平均准确率为89.38%,比GraphRNA（Graph Recurrent Networks with Attributed random walks）高出了2.02个百分点,比经典工作DeepWalk高出了21.12个百分点;同时,对比不同随机游走方法发现,提高对节点关联有促进作用的属性的采样概率可以增加采样序列所含信息。     

基于深度神经网络和门控循环单元的动态图表示学习方法

学习图中节点的潜在向量表示是一项重要且普遍存在的任务,旨在捕捉图中节点的各种属性。大量工作证明静态图表示已经能够学习到节点的部分信息,然而,真实世界的图是随着时间的推移而演变的。为了解决多数动态网络算法不能有效保留节点邻域结构和时态信息的问题,提出了基于深度神经网络（DNN）和门控循环单元（GRU）的动态网络表示学习方法DynAEGRU。该方法以自编码器作为框架,其中的编码器首先用DNN聚集邻域信息以得到低维特征向量,然后使用GRU网络提取节点时态信息,最后用解码器重构邻接矩阵并将其与真实图对比来构建损失。通过与几种静态图和动态图表示学习算法在3个数据集上进行实验分析,结果表明DynAEGRU具有较好的性能增益。     

基于用户传播网络与消息内容融合的谣言检测模型

针对社交媒体平台上消息内容普遍很短、传播结构中存在大量空转发、用户角色与内容间的失配等条件约束,提出了一种基于传播网络中的用户属性信息和消息内容的谣言检测模型GMB_GMU。首先以用户属性为节点、传播链为边构建用户传播网络,并引入图注意力网络（GAT）得到用户属性的增强表示;同时,基于此用户传播网络,利用node2vec得到用户的结构表征,并使用互注意机制对其进行增强。另外,引入BERT建立源帖内容表征。最后,利用多模态门控单元（GMU）对用户属性表征、结构表征和源帖内容表征进行融合,从而得到消息的最终表征。实验结果表明,GMB_GMU模型在公开的Weibo数据上的准确率达到0.952,能够有效识别谣言事件,效果明显优于基于循环神经网络（RNN）和其他神经网络基准模型的传播算法。     

图数据发布隐私保护的聚类匿名方法

社交网络中积累的海量信息构成一类图大数据,为防范隐私泄露,一般在发布此类数据时需要做匿名化处理.针对现有匿名方案难以防范同时以结构和属性信息为背景知识的攻击的不足,研究一种基于节点连接结构和属性值的属性图聚类匿名化方法,利用属性图表示社交网络数据,综合根据节点间的结构和属性相似度,将图中所有节点聚类成一些包含节点个数不小于k的超点,特别针对各超点进行匿名化处理.该方法中,超点的子图隐匿和属性概化可以分别防范一切基于结构和属性背景知识的识别攻击.另外,聚类过程平衡了节点间的连接紧密性和属性值相近性,有利于减小结构和属性的总体信息损失值,较好地维持数据的可用性.实验结果表明了该方法在实现算法功能和减少信息损失方面的有效性.     

融合概念和属性信息的领域知识图谱补全方法

针对领域知识图谱具有严格的模式层和丰富的属性信息的特点,提出一种融合概念和属性信息的领域知识图谱补全方法.首先对领域知识图谱模式层中的概念使用可建模语义分层结构的HAKE模型进行嵌入表示,建立基于概念的实例向量表示;其次对数据层的实例三元组和属性三元组进行区分,通过注意力机制对实例的属性和概念进行融合,建立基于属性的实例向量表示;最后对基于概念和基于属性的实例向量表示进行联合训练以实现对实例三元组的评分.使用基于DWY100K数据集构建的知识图谱、MED-BBK-9K 医疗知识图谱和根据某钢铁企业设备故障诊断数据构建的知识图谱进行实验,结果表明所提出方法在领域知识图谱补全中的性能优于现有知识图谱补全方法.     

Centrality Based Visualization of Small World Graphs

Current graph drawing algorithms enable the creation of two dimensional node‐link diagrams of huge graphs. However, for graphs with low diameter (of which “small world” graphs are a subset) these techniques begin to break down visually even when the graph has only a few hundred nodes. Typical algorithms produce images where nodes clump together in the center of the screen, making it hard to discern structure and follow paths. This paper describes a solution to this problem, which uses a global edge metric to determine a subset of edges that capture the graph's intrinsic clustering structure. This structure is then used to create an embedding of the graph, after which the remaining edges are added back in. We demonstrate applications of this technique to a number of real world examples.     

融合知识图谱和轻量图卷积网络的推荐算法

针对知识图谱推荐算法用户端和项目端建模程度不均且模型复杂度较高等问题, 提出融合知识图谱和轻量图卷积网络的推荐算法. 在用户端, 利用用户相似性生成邻居集合, 将用户及其相似用户的交互记录在知识图谱上多次迭代传播, 增强用户特征表示. 在项目端, 将知识图谱中实体嵌入传播, 挖掘与用户喜好相关的项目信息; 接着, 利用轻量图卷积网络聚合邻域特征获得用户和项目的特征表示, 同时采用注意力机制将邻域权重融入实体, 增强节点的嵌入表示; 最后, 预测用户和项目之间的评分. 实验表明, 在Book-Crossing数据集上, 相较于最优基线, AUC和ACC分别提高了1.8%和2.3%. 在Yelp2018数据集上, AUC和ACC分别提高了1.2%和1.4%. 结果证明, 该模型与其他基准模型相比有较好的推荐性能.