Graph Visualization Techniques Research Articles

Complex business ecosystem intelligence using AI-powered visual analytics

Business ecosystems are complex, dynamic systems characterized by a multitude of entities, including companies, ventures, and technologies, as well as activities and trends. Understanding the state of business ecosystems is an increasingly critical strategic imperative for many decision makers, but it is a resource-intensive activity as relevant information sources are dispersed, often highly unstructured, and not integrated or curated to deliver actionable insights. In this research, we present the design and implementation of an interactive visual analytic system that integrates artificial intelligence and graph visualization techniques to augment decision makers’ understanding of the complex public narrative associated with business ecosystems entities. Our system is driven by a real-time content engine of 100,000+ global data sources including press releases, news articles, industry reports, analyst blogs in multiple languages organized across several domain-specific repositories. Following a user-specified query, the engine extracts both domain-agnostic and domain-specific entities and concepts for each document in the result set. We then model and visualize the resulting data as a dynamic, multipartite network and implement graph pruning algorithms and interactive data controls to enable users to interactively explore and discover the underlying business ecosystem from multiple perspectives. We illustrate and discuss the value of our system using representative use cases. Our study makes multiple contributions to visual decision support theory and practice, including mining unstructured data, constructing and interacting with knowledge graphs, and designing visual analytic tools for ecosystem intelligence. We conclude the study with implications and future research opportunities.

GeoGraphViz: Geographically constrained 3D force‐directed graph for knowledge graph visualization

AbstractKnowledge graphs are a key technique for linking and integrating cross‐domain data, concepts, tools, and knowledge to enable data‐driven analytics. As much of the world's data have become massive in size, visualizing graph entities and their interrelationships intuitively and interactively has become a crucial task for ingesting and better utilizing graph content to support semantic reasoning, discovering hidden knowledge discovering, and better scientific understanding of geophysical and social phenomena. Despite the fact that many such phenomena (e.g., disasters) have clear spatial footprints and geographic properties, their location information is considered only as a textual label in existing graph visualization tools, limiting their capability to reveal the geospatial distribution patterns of the graph nodes. In addition, most graph visualization techniques rely on 2D graph visualization, which constrains the dimensions of information that can be presented and lacks support for graph structure examination from multiple angles. To tackle the above challenges, we developed a novel 3D map‐based graph visualization algorithm to enable interactive exploration of graph content and patterns in a spatially explicit manner. The algorithm extends a 3D force directed graph by integrating a web map, an additional geolocational force, and a force balancing variable that allows for the dynamic adjustment of the 3D graph structure and layout. This mechanism helps create a balanced graph view between the semantic forces among the graph nodes and the attractive force from a geolocation to a graph node. Our solution offers a new perspective in visualizing and understanding spatial entities and events in a knowledge graph.

Graph Interpretation, Summarization and Visualization Techniques: A Review and Open Research Issues

Graph Interpretation, Summarization and Visualization Techniques: A Review and Open Research Issues

TrammelGraph: visual graph abstraction for comparison

Network data, represented by graph-based structures, are used in a variety of applications such as social networks and disease complication networks. A crucial task in many applications is graph comparison, with the goal of understanding the structural differences between pairs of graphs. However, traditional graph visualization techniques, node-link diagrams, and adjacency matrices are not intrinsically designed for comparison tasks. In this paper, we present TrammelGraph, a novel visual graph abstraction technique for graph comparison. TrammelGraph uses graph embedding and node alignment methods to create a map-based representation that eliminates visual clutter and thus facilities graph comparison. The design results in a planar graph with a regularly-spaced, crossing-free layout that helps users identify high-level topological information for graph comparison tasks. To evaluate the effectiveness of TrammelGraph, we conducted a controlled user study with 20 participants that compares TrammelGraph with node-link diagrams and adjacency matrices on a variety of common graph comparison tasks. We also conducted an expert interview with a physician using the MIMIC dataset. Both the quantitative and qualitative results from the study and interview highlighted the key strengths of TrammelGraph as a tool for visual graph comparison.

Toward automatic comparison of visualization techniques: Application to graph visualization

Many end-user evaluations of data visualization techniques have been run during the last decades. Their results are cornerstones to build efficient visualization systems. However, designing such an evaluation is always complex and time-consuming and may end in a lack of statistical evidence and reproducibility. We believe that modern and efficient computer vision techniques, such as deep convolutional neural networks (CNNs), may help visualization researchers to build and/or adjust their evaluation hypothesis. The basis of our idea is to train machine learning models on several visualization techniques to solve a specific task. Our assumption is that it is possible to compare the efficiency of visualization techniques based on the performance of their corresponding model. As current machine learning models are not able to strictly reflect human capabilities, including their imperfections, such results should be interpreted with caution. However, we think that using machine learning-based pre-evaluation, as a pre-process of standard user evaluations, should help researchers to perform a more exhaustive study of their design space. Thus, it should improve their final user evaluation by providing it better test cases. In this paper, we present the results of two experiments we have conducted to assess how correlated the performance of users and computer vision techniques can be. That study compares two mainstream graph visualization techniques: node-link (NL) and adjacency-matrix (AM) diagrams. Using two well-known deep convolutional neural networks, we partially reproduced user evaluations from Ghoniem et al. and from Okoe et al.. These experiments showed that some user evaluation results can be reproduced automatically.

Visual querying and analysis of temporal fiscal networks

We present TeFNet, a new system for the visual analysis of temporal networks in the fiscal domain, aimed to contrast tax evasion, fiscal frauds, and money laundering. The design of TeFNet has been driven by domain experts (tax officers) and the system is currently used by the Italian revenue agency, Agenzia delle Entrate. TeFNet is based on a powerful visual query language that allows users to easily define suspicious patterns involving temporal constraints. To efficiently execute graph pattern matching algorithms on large networks, TeFNet exploits modern graph database technologies. Besides its visual query language, TeFNet is equipped with graph visualization techniques to quickly convey time-varying information during an interactive exploration of the subgraphs that match a desired pattern. Both the query language and the visualization techniques rely on a suitable timeline approach, which maps the time dimension to the space dimension. The system has been tested in a real working environment. We demonstrate its effectiveness by discussing use cases on real suspicious patterns and the results of experiments conducted with expert tax officers on real data. Furthermore, we performed a qualitative evaluation to get additional insights about advantages and limits of our system.

MonitorApp: a web tool to analyze and visualize pollution data detected by an electronic nose

The analysis of air quality data may reveal the quality of life and can prevent dangers for the citizen health. Assuming that some chemical compounds in the air produce a bad smell, people may detect that something is going wrong acting as sensors that alerts potential risks. This work presents a visual analytics approach to support air quality experts in the analysis of data produced by electronic nose devices. The approach consists in setting workflows to manage and transform raw data offering clustering and visualization techniques to analyze such information. The analysis is supported by calendar, map and line graph visualization techniques also maneuvering the clustering attributes. The interactive map is used to show the position of monitoring stations in order to support making hypothesis related to the data source locations.

Automatyzacja wizualizacji grafowej synonimów dla potrzeb dydaktyki języków obcych (na przykładzie przymiotników rosyjskich z prefiksem „без-”/„бес-”)

The article presents methods for application of lexicographic data for the purposes of foreign language teaching and linguistic studies. The automated graph visualization technique is introduced. Both the retrieval and the visualization of linguistic data in form of directed graphs have been conducted with the use of available computer software. The material on which the above-mentioned operations have been performed comprises the group of Russian adjectives with без- / бес- prefixes and the synonymic chains initiated by them. The automated extraction with the use of regular expressions, the steps leading to the edition of the generated database and the import of the final data base into a visualization software are described in the article. The grammatical and semantic characteristics of the sample group of lexemes are not the focus of this article, although further analysis of the selected vocabulary units is planned to be conducted in future studies.

Juniper: A Tree+ Table Approach to Multivariate Graph Visualization.

Analyzing large, multivariate graphs is an important problem in many domains, yet such graphs are challenging to visualize. In this paper, we introduce a novel, scalable, tree+table multivariate graph visualization technique, which makes many tasks related to multivariate graph analysis easier to achieve. The core principle we follow is to selectively query for nodes or subgraphs of interest and visualize these subgraphs as a spanning tree of the graph. The tree is laid out linearly, which enables us to juxtapose the nodes with a table visualization where diverse attributes can be shown. We also use this table as an adjacency matrix, so that the resulting technique is a hybrid node-link/adjacency matrix technique. We implement this concept in Juniper and complement it with a set of interaction techniques that enable analysts to dynamically grow, restructure, and aggregate the tree, as well as change the layout or show paths between nodes. We demonstrate the utility of our tool in usage scenarios for different multivariate networks: a bipartite network of scholars, papers, and citation metrics and a multitype network of story characters, places, books, etc.

Towards Visualizing Big Data with Large-Scale Edge Constraint Graph Drawing

Abstract Visualization plays an important role in enabling understanding of big data. Graphs are crucial tools for visual analytics of big data networks such as social, biological, traffic and security networks. Graph drawing has been intensively researched to enhance aesthetic features (i.e., layouts, symmetry, cross-free edges). Early physic-inspired techniques have focused on synthetic abstract graphs whose weights/distances of the edges are often ignored or assumed equal. Although recent approaches have been extended to sophisticated realistic networks, most are not designed to address very large-scale weighted graphs, which are important for visual analyses. The difficulty lies in the fact that the drawing process, governed by these physical properties, oscillates in large graphs and conflicts with specified distances leading to poor visual results. Our research attempts to alleviate these obstacles. This paper presents a simple graph visualization technique that aims to efficiently draw aesthetically pleasing large-scale straight-line weighted edge graphs. Our approach uses relevant physic-inspired techniques to promote aesthetic graphs and proposes a weak constraint-based approach to handle large-scale computing and competing goals to satisfy both weight requirements and aesthetic properties. The paper describes the approach along with experiments on both synthetic and real large-scale weighted graphs including that of over 10,000 nodes and comparisons with state-of-the-art approaches. The results obtained show enhanced and promising outcomes toward a general-purpose graph drawing technique for both big synthetic and real network data analytics.

Climate change effects and their interactions: An analysis aiming at policy implications

In this study we provide a computerized graph structure for synthesizing and displaying the data on a region’s ecosystem-economic system. By applying Mathematica-based graph modeling we create a causal network of the synergistic impact mechanism among certain climate related factors. Our computational approach identifies a climate factor that affects most immediately or most strongly the others. Important factors are indicated through the use of graph theoretical tools. Our graph-based approach and its computational aspects allow for factor ranking(s) according to their importance to the network both numerically and visually, for certain settlement types. Our contribution provides quantitative estimates of impacts and adaptation potentials of five potential effects of climate change (migration, flooding-landslides-fire, air and water pollution, human health and energy-water-other resources) which play a substantial role at the synergistic impact mechanism. By using graph visualization techniques, the structure of the synergistic impact mechanism is self-evident. Specifically, graph layouts are created to detect i) the causal relationships of the synergistic mechanism under study ii) the most influential factor(s) in the synergistic mechanism and iii) classify the factor’s roles (based on the degree of their impact) within the coping mechanism. Highlighting graph elements let information for policy implications stand out.

The dynamic graph wall: visualizing evolving graphs with multiple visual metaphors

Visualizing dynamic graphs is challenging due to the many data dimensions to be displayed such as graph vertices and edges with their attached weights or attributes and the additional time dimension. Moreover, edge directions with multiplicities and the graph topology are also important inherent features. However, in many dynamic graph visualization techniques each graph in a sequence is treated the same way, i.e., it is visually encoded in the same visual metaphor or even in the same layout. This visualization strategy can be problematic if the graphs are changing topologically over time, i.e., if a sparse graph becomes denser and denser over time or a star pattern is changing into a dense cluster of connected vertices. Such a dynamic graph data scenario demands for a visualization approach which is able to adapt the applied visual metaphor to each graph separately. In this paper we describe the dynamic graph wall to solve this problem by using multiple visual metaphors for dynamic graphs which are computed automatically by algorithms analysing each individual graph based on a given repertoire of graph features. The biggest issue in this technique for the graph dynamics, however, is the preservation of the viewer’s mental map at metaphor changes, i.e., to guide him through the graph changes with the goal to explore the data for time-varying patterns. To reach this goal we support the analyst by an interactive highlighting feature but we also display graphs in comparative metaphor rows to visually investigate the commonalities and differences over time. Graphical Abstract text

EgoSlider: Visual Analysis of Egocentric Network Evolution.

Ego-network, which represents relationships between a specific individual, i.e., the ego, and people connected to it, i.e., alters, is a critical target to study in social network analysis. Evolutionary patterns of ego-networks along time provide huge insights to many domains such as sociology, anthropology, and psychology. However, the analysis of dynamic ego-networks remains challenging due to its complicated time-varying graph structures, for example: alters come and leave, ties grow stronger and fade away, and alter communities merge and split. Most of the existing dynamic graph visualization techniques mainly focus on topological changes of the entire network, which is not adequate for egocentric analytical tasks. In this paper, we present egoSlider, a visual analysis system for exploring and comparing dynamic ego-networks. egoSlider provides a holistic picture of the data through multiple interactively coordinated views, revealing ego-network evolutionary patterns at three different layers: a macroscopic level for summarizing the entire ego-network data, a mesoscopic level for overviewing specific individuals' ego-network evolutions, and a microscopic level for displaying detailed temporal information of egos and their alters. We demonstrate the effectiveness of egoSlider with a usage scenario with the DBLP publication records. Also, a controlled user study indicates that in general egoSlider outperforms a baseline visualization of dynamic networks for completing egocentric analytical tasks.

Key-Node-Separated Graph Clustering and Layouts for Human Relationship Graph Visualization.

Many graph-drawing methods apply node-clustering techniques based on the density of edges to find tightly connected subgraphs and then hierarchically visualize the clustered graphs. However, users may want to focus on important nodes and their connections to groups of other nodes for some applications. For this purpose, it is effective to separately visualize the key nodes detected based on adjacency and attributes of the nodes. This article presents a graph visualization technique for attribute-embedded graphs that applies a graph-clustering algorithm that accounts for the combination of connections and attributes. The graph clustering step divides the nodes according to the commonality of connected nodes and similarity of feature value vectors. It then calculates the distances between arbitrary pairs of clusters according to the number of connecting edges and the similarity of feature value vectors and finally places the clusters based on the distances. Consequently, the technique separates important nodes that have connections to multiple large clusters and improves the visibility of such nodes' connections. To test this technique, this article presents examples with human relationship graph datasets, including a coauthorship and Twitter communication network dataset.

Social graph visualization techniques for public data

Social graph visualization techniques for public data

Visual Exploration of Sparse Traffic Trajectory Data.

In this paper, we present a visual analysis system to explore sparse traffic trajectory data recorded by transportation cells. Such data contains the movements of nearly all moving vehicles on the major roads of a city. Therefore it is very suitable for macro-traffic analysis. However, the vehicle movements are recorded only when they pass through the cells. The exact tracks between two consecutive cells are unknown. To deal with such uncertainties, we first design a local animation, showing the vehicle movements only in the vicinity of cells. Besides, we ignore the micro-behaviors of individual vehicles, and focus on the macro-traffic patterns. We apply existing trajectory aggregation techniques to the dataset, studying cell status pattern and inter-cell flow pattern. Beyond that, we propose to study the correlation between these two patterns with dynamic graph visualization techniques. It allows us to check how traffic congestion on one cell is correlated with traffic flows on neighbouring links, and with route selection in its neighbourhood. Case studies show the effectiveness of our system.

GLO-STIX: Graph-Level Operations for Specifying Techniques and Interactive eXploration.

The field of graph visualization has produced a wealth of visualization techniques for accomplishing a variety of analysis tasks. Therefore analysts often rely on a suite of different techniques, and visual graph analysis application builders strive to provide this breadth of techniques. To provide a holistic model for specifying network visualization techniques (as opposed to considering each technique in isolation) we present the Graph-Level Operations (GLO) model. We describe a method for identifying GLOs and apply it to identify five classes of GLOs, which can be flexibly combined to re-create six canonical graph visualization techniques. We discuss advantages of the GLO model, including potentially discovering new, effective network visualization techniques and easing the engineering challenges of building multi-technique graph visualization applications. Finally, we implement the GLOs that we identified into the GLO-STIX prototype system that enables an analyst to interactively explore a graph by applying GLOs.

An advanced pre-positioning method for the force-directed graph visualization based on pagerank algorithm

Graph visualization techniques aim to help users obtain deep insights about the structure and evolution of graphs through minimizing visual confusion and real-time updating of the drawing. This demands high effectiveness of the drawing algorithms to produce layout of graphs with few edge crossings. However, the classical force-directed algorithms fail to satisfy this requirement. In this paper, we solve this problem by using a novel and heuristic pre-positioning method based on a variant of the PageRank algorithm. In our algorithmic framework, five models are proposed to calculate the importance degree of each vertex, the relation matrix of the vertices, the layer each vertex belongs to, the available placing coordinates and the matching coefficient between a vertex and a position. After the pre-positioning procedure of each layer, the classical force-directed algorithm is used to calculate the final position of each vertex. Experimental results have shown that our approach can reduce the computing time as well as the number of edge crossings.

Characterization of the peer review network at the Center for Scientific Review, National Institutes of Health.

The National Institutes of Health (NIH) is the largest source of funding for biomedical research in the world. This funding is largely effected through a competitive grants process. Each year the Center for Scientific Review (CSR) at NIH manages the evaluation, by peer review, of more than 55,000 grant applications. A relevant management question is how this scientific evaluation system, supported by finite resources, could be continuously evaluated and improved for maximal benefit to the scientific community and the taxpaying public. Towards this purpose, we have created the first system-level description of peer review at CSR by applying text analysis, bibliometric, and graph visualization techniques to administrative records. We identify otherwise latent relationships across scientific clusters, which in turn suggest opportunities for structural reorganization of the system based on expert evaluation. Such studies support the creation of monitoring tools and provide transparency and knowledge to stakeholders.

Effective email network visualization techniques by means of user behaviors

In email or twitter networks, mining interaction behaviors of actors is a useful task in viral marketing or targeted advertisement. In this work, we define two interaction behaviors. One is engagingness representing the ability of soliciting active response for an initiated message. The other is responsiveness that indicates the ability of responding to an incoming message in information exchange networks. To study such behaviors, we make use of Enron email data set that are so far the only known publicly available information exchange data with messages assigned with specific senders and recipients. We also conduct data preprocessing on the email data and establish links between emails and their replies. Then, we propose quantitative behavior models for systematically measuring each user's engagingness and responsiveness scores in the email network. Further we present a graph visualization technique in order to visualize information exchange networks by means of the concept of behaviors and community structures. In our empirical study, we compare the proposed behavior models, and introduce several properties of behaviors out which we found with the Enron email data set. In addition, our visualization technique will be better able to figure out the underlying characteristics of information exchange networks based on our behavior models.