This paper describes the process of developing data visualisations to enhance a commercial software platform for combating insider threat, whose existing UI, while perfectly functional, was limited in its ability to allow analysts to easily spot the patterns and outliers that visualisation naturally reveals. We describe the design and development process, proceeding from initial tasks/requirements gathering, understanding the platform’s data formats, the rationale behind the visualisations’ design, and then refining the prototype through gathering feedback from representative domain experts who are also current users of the software. Through a number of example scenarios, we show that the visualisation can support the identified tasks and aid analysts in discovering and understanding potentially risky insider activity within a large user base.

Abstract This paper reports the design and comparison of three visualizations to represent the structure and content within arguments. Arguments are artifacts of reasoning widely used across domains such as education, policy making, and science. Anargumentis made up of sequences of statements (premises) which can support or contradict each other, individually or in groups through Boolean operators. Understanding the resulting hierarchical structure of arguments while being able to read the arguments' text poses problems related to overview, detail, and navigation. Based on interviews with argument analysts we iteratively designed three techniques, each using combinations of tree visualizations (sunburst, icicle), content display (in‐situ, tooltip) and interactive navigation. Structured discussions with the analysts show benefits of each these techniques; for example, sunburst being good in presenting overview but showing arguments in‐situ is better than pop‐ups. A controlleduser study with 21 participants and three tasks shows complementary evidence suggesting that a sunburst with pop‐up for the content is the best trade‐off solution. Our results can inform visualizations within existing argument visualization tools and increase the visibility of ‘novel‐and‐effective’ visualizations in the argument visualization community.

This paper reports on a formal user study on visual encodings of networks with multiple edge types in adjacency matrices. Our tasks and conditions were inspired by real problems in computational biology. We focus on encodings in adjacency matrices, selecting four designs from a potentially huge design space of visual encodings. We then settle on three visual variables to evaluate in a crowdsourcing study with 159 participants: orientation, position and colour. The best encodings were integrated into a visual analytics tool for inferring dynamic Bayesian networks and evaluated by computational biologists for additional evidence. We found that the encodings performed differently depending on the task, however, colour was found to help in all tasks except when trying to find the edge with the largest number of edge types. Orientation generally outperformed position in all of our tasks.

We address the problem of exploring, combining, and comparing large collections of scored, directed networks for understanding inferred Bayesian networks used in biology. In this field, heuristic algorithms explore the space of possible network solutions, sampling this space based on algorithm parameters and a network score that encodes the statistical fit to the data. The goal of the analyst is to guide the heuristic search and decide how to determine a final consensus network structure, usually by selecting the top-scoring network or constructing the consensus network from a collection of high-scoring networks. BayesPiles, our visualisation tool, helps with understanding the structure of the solution space and supporting the construction of a final consensus network that is representative of the underlying dataset. BayesPiles builds upon and extends MultiPiles to meet our domain requirements. We developed BayesPiles in conjunction with computational biologists who have used this tool on datasets used in their research. The biologists found our solution provides them with new insights and helps them achieve results that are representative of the underlying data.

Abstract Categorising tasks is a common pursuit in the visualisation research community, with a wide variety of taxonomies, typologies, design spaces, and frameworks having been developed over the last three decades. While these classifications are universally purported to be useful in both the design and evaluation processes and in guiding future research, remarkably little attention has been paid to how these frameworks have—and can be—constructed and evaluated. In this paper we review the task classification literature and report on current practices in construction and evaluation. We consider the stages of task classification construction and identify the associated threats to validity arising at each stage and in response to the different methods employed. We provide guidance on suitable validation approaches in order to mitigate these threats. We also consider the appropriateness of evaluation strategies according to the different aspects of the classification which they evaluate. In so doing, we seek to provide guidance for developers of classifications in determining appropriate construction and evaluation strategies when developing a classification, and also for those selecting between competing classifications for use in the design and evaluation processes.

In this paper, we explore graph comics as a medium to communicate changes in dynamic networks. While previous re- search has focused on visualizing dynamic networks for data exploration, we want to see if we can take advantage of the visual expressiveness and familiarity of comics to present and explain temporal changes in networks to an audience. To understand the potential of comics as a storytelling medium, we first created a variety of comics during a 3 month structured design process, involving domain experts from public education and neuroscience. This process led to the definition of 8 design factors for creating graph comics and propose design solutions for each. Results from a qualitative study suggest that a general audience is quickly able understand complex temporal changes through graph comics, provided with minimal textual annotations and no training.

By extending and instantiating an existing formal task framework, we define a task taxonomy and task design space for temporal graph visualisation. We discuss the process involved in their generation, and describe how the design space can be 'sliced and diced' into multiple overlapping task categories, requiring distinct visual techniques for their support. The approach addresses deficiencies in the task literature, offering domain independence, greater task coverage, and unambiguous task specification. The taxonomy and design space capture tasks for temporal graphs, and also static graphs, multivariate graphs, and graph comparison, and will be of value in the design and evaluation of temporal graph visualisation systems.

BACKGROUND: Plant breeders use an increasingly diverse range of data types to identify lines with desirable characteristics suitable to be taken forward in plant breeding programmes. There are a number of key morphological and physiological traits, such as disease resistance and yield that need to be maintained and improved upon if a commercial variety is to be successful. Computational tools that provide the ability to integrate and visualize this data with pedigree structure, will enable breeders to make better decisions on the lines that are used in crossings to meet both the demands for increased yield/production and adaptation to climate change. RESULTS: We have used a large and unique set of experimental barley (H. vulgare) data to develop a prototype pedigree visualization system. We then used this prototype to perform a subjective user evaluation with domain experts to guide and direct the development of an interactive pedigree visualization tool called Helium. CONCLUSIONS: We show that Helium allows users to easily integrate a number of data types along with large plant pedigrees to offer an integrated environment in which they can explore pedigree data. We have also verified that users were happy with the abstract representation of pedigrees that we have used in our visualization tool.

Monitoring log files for network intrusions is unwieldy. To build a mental model of the log, an analyst is required to recognise continuous timelines and attack patterns from a dataset that is essentially limited to an ordered list of events. Information Visualization techniques arrange data into directly perceivable visual patterns that may alleviate some overheads associated with interpreting these datasets and improve the ability of users, especially those in resource-stretched Small and Medium sized Businesses (SMBs), to make sense of activity patterns in Intrusion Detection System (IDS) event logs. To this end, we discuss existing network security visualizations for IDS logs and after examining the strengths and drawbacks of those applications we have prototyped a visualization tool, Pianola, that arranges events on multiple timelines to reveal patterns both in time and across a network. The tool was evaluated against the traditional use of command-line interface (CLI)-based tools for analyzing network security events and displayed significant improvements in both recognition and detection of attacks and reduction in the users' subjective workload, measured using the NASA Task Load index (TLX).

BACKGROUND: High throughput gene expression time-course experiments provide a perspective on biological functioning recognized as having huge value for the diagnosis, treatment, and prevention of diseases. There are however significant challenges to properly exploiting this data due to its massive scale and complexity. In particular, existing techniques are found to be ill suited to finding patterns of changing activity over a limited interval of an experiments time frame. The Time-Series Explorer (TSE) was developed to overcome this limitation by allowing users to explore their data by controlling an animated scatter-plot view. MaTSE improves and extends TSE by allowing users to visualize data with missing values, cross reference multiple conditions, highlight gene groupings, and collaborate by sharing their findings. RESULTS: MaTSE was developed using an iterative software development cycle that involved a high level of user feedback and evaluation. The resulting software combines a variety of visualization and interaction techniques which work together to allow biologists to explore their data and reveal temporal patterns of gene activity. These include a scatter-plot that can be animated to view different temporal intervals of the data, a multiple coordinated view framework to support the cross reference of multiple experimental conditions, a novel method for highlighting overlapping groups in the scatter-plot, and a pattern browser component that can be used with scatter-plot box queries to support cooperative visualization. A final evaluation demonstrated the tools effectiveness in allowing users to find unexpected temporal patterns and the benefits of functionality such as the overlay of gene groupings and the ability to store patterns. CONCLUSIONS: We have developed a new exploratory analysis tool, MaTSE, that allows users to find unexpected patterns of temporal activity in gene expression time-series data. Overall, the study acted well to demonstrate the benefits of an iterative software development life cycle and allowed us to investigate some visualization problems that are likely to be common in the field of bioinformatics. The subjects involved in the final evaluation were positive about the potential of MaTSE to help them find unexpected patterns in their data and characterized MaTSE as an exploratory tool valuable for hypothesis generation and the creation of new biological knowledge.

We describe a visual data cleansing application for pedigree genotype data, which is used to redeem otherwise unusable pedigree data sets. Biologists and bioinformaticians dynamically and iteratively mask pieces of information from a dirty data set and graduate towards a usable cleaned version of the data, which can then be saved and used in ongoing biological analyses. Cleansing of such data is complicated over and above simple error cleaning in that change or masking of the pedigree structure may shift errors to new parts of the pedigree. Thus a branching history of data manipulations is kept to allow users to restore the data and visualisation to any of the previous states it has travelled through.

This Supplement to BMC Bioinformatics contains a collection of extended articles providing highlights from the 1st IEEE Symposium on Biological Data Visualization (BioVis 2011, http://www.biovis.net) held on October 23-24, 2011, in conjunction with IEEE Visweek in Providence, RI. The rapidly expanding field of biology creates enormous challenges for computational visualization techniques for enabling researchers to gain insight from large and highly complex data sets. The goal of the BioVis 2011 symposium was to create the premier international and interdisciplinary event for all aspects of visualization in biology. The symposium brought together researchers from the visualization, bioinformatics, and biology communities with the purpose of educating, inspiring, and engaging visualization researchers in problems in biological data visualization, as well as bioinformatics and biology researchers in state-of-the-art visualization research. The symposium also served as a platform for researchers in biology and bioinformatics to share pressing visualization challenges and potential solutions in their fields, to initiate interdisciplinary collaborations and to provide an outlet and training ground for young and freshly minted visualization researchers with a keen interest in problems of biology. For the BioVis 2011 Symposium, a new data visualization/analysis contest was inaugurated. The main goals of the contest were: the development of a better-informed visualization community, provided with deeper domain-specific intuition into the actual issues of interest to the biology and bioinformatics communities; a better-tooled biological community, provided with enhanced application software specifically adjusted to meet their analysis needs; and, finally, a mechanism to strongly promote fundable peer collaborations between visualization and bioinformatics/biology researchers. This contest was focused on specific biological problem domains, based on realistic domain data and domain questions. For BioVis 2011, the contest involved the analysis of expression Quantitative Trait Locus (eQTL) data. Judging for the contest entries was conducted by a panel of experts, including members from both the bioinformatics/biology and the visualization communities. The procedure for selecting the articles for this Supplement was as follows. A first selection of best papers from the twenty paper contributions to BioVis 2011 was made by the paper chairs, based on the reviewer scores and recommendations and the authors were requested to submit an extended version of their paper with at least 30% additional, unpublished, material. In addition, one paper was invited to discuss the BioVis 2011 contest, summarizing the problem and the contest submissions and discussing the lessons learned, for example, what worked, what didn't, and what seemed to be the biggest challenges in coming to a solution. These papers went through a new reviewing round with two independent reviewers. In agreement with the BMC peer review guidelines, the supplement editors did not review or make acceptance decisions about any manuscript to which they had contributed. Following the reviewing process papers were either accepted after taking into account the recommendations by the reviewers or rejected. As a result, seven papers have been accepted for this Supplement. The articles in this Supplement cover a wide spectrum of challenging problems in biological data visualization and their solutions. Heinrich et al. describe the interactive Hierarchical Aggregation Table (iHAT), which facilitates the visualization of multiple sequence alignments, associated metadata, and hierarchical clusterings. Smith et al. present RuleBender, a novel visualization system for the integrated visualization, modeling and simulation of rule-based intracellular biochemistry. In the same spirit, Strobelt et al. present HiTSEE (High-Throughput Screening Exploration Environment), a visualization tool for the analysis of large chemical screens used to examine biochemical processes. Paterson et al. describe VIPER (Visual Pedigree Explorer), a software tool for pedigree visualization that integrates an inheritance-checking algorithm with a novel space-efficient pedigree visualization. Livengood et al. present a system that enables interactive comparative visualization and analytics of metabolomics data obtained by two-dimensional gas chromatography-mass spectrometry. Mayerich et al. present NetMets, a method for quantifying and visualizing errors in biological network segmentation. Finally, Bartlett et al. present and discuss the eQTL biological data visualization challenge conducted as the BioVis 2011 contest. This is a tremendously important biological grand challenge domain with no extant solutions. The authors provide a thought-provoking perspective on the complex relations between the biology, bioinformatics, and visualization domains. They discuss the contest entries submitted to BioVis 2011, present the lessons learned, and conclude with open questions for the future. We hope that the collection of papers in this Supplement will enlighten the reader about the intense activity in the challenging field of biological data visualization, and will encourage and stimulate the interaction and collaboration between researchers from the biology, bioinformatics, and visualization communities.

BACKGROUND: Pedigree genotype datasets are used for analysing genetic inheritance and to map genetic markers and traits. Such datasets consist of hundreds of related animals genotyped for thousands of genetic markers and invariably contain multiple errors in both the pedigree structure and in the associated individual genotype data. These errors manifest as apparent inheritance inconsistencies in the pedigree, and invalidate analyses of marker inheritance patterns across the dataset. Cleaning raw datasets of bad data points (incorrect pedigree relationships, unreliable marker assays, suspect samples, bad genotype results etc.) requires expert exploration of the patterns of exposed inconsistencies in the context of the inheritance pedigree. In order to assist this process we are developing VIPER (Visual Pedigree Explorer), a software tool that integrates an inheritance-checking algorithm with a novel space-efficient pedigree visualisation, so that reported inheritance inconsistencies are overlaid on an interactive, navigable representation of the pedigree structure. METHODS AND RESULTS: This paper describes an evaluation of how VIPER displays the different scales and types of dataset that occur experimentally, with a description of how VIPER's display interface and functionality meet the challenges presented by such data. We examine a range of possible error types found in real and simulated pedigree genotype datasets, demonstrating how these errors are exposed and explored using the VIPER interface and we evaluate the utility and usability of the interface to the domain expert.Evaluation was performed as a two stage process with the assistance of domain experts (geneticists). The initial evaluation drove the iterative implementation of further features in the software prototype, as required by the users, prior to a final functional evaluation of the pedigree display for exploring the various error types, data scales and structures. CONCLUSIONS: The VIPER display was shown to effectively expose the range of errors found in experimental genotyped pedigrees, allowing users to explore the underlying causes of reported inheritance inconsistencies. This interface will provide the basis for a full data cleaning tool that will allow the user to remove isolated bad data points, and reversibly test the effect of removing suspect genotypes and pedigree relationships.

While much work has been done in the area of visualization for analysis of graphs, relatively little research exists into how best to use visualization for comparing graphs. We have developed a suite of general graph comparison questions that can be tailored to specific data sets, and compared the use of superimposed and juxtaposed views of graph matrices on an example data set. Our observations indicate that combined views are more useful in comparing general graphs, allowing for greater user accuracy in determining differences and their effects.

Abstract Genetic analysis of a breeding animal population involves determining the inheritance pattern of genotypes for multiple genetic markers across the individuals in the population pedigree structure. However, experimental pedigree genotype data invariably contains errors in both the pedigree structure and in the associated individual genotypes, introducing inconsistencies into the dataset, rendering them useless for further analysis. The resolution of these errors requires consideration of genotype inheritance patterns in the context of the pedigree structure. Existing pedigree visualisations are typically more suited to human pedigrees and are less suitable for large complex animal pedigrees which may exhibit cross generational inbreeding. Similarly, table‐based viewers of genotype marker data can highlight where errors become apparent but lack the functionality and interactive visual feedback to allow users to locate the origin of errors within the pedigree. In this paper, we detail a design study steered by biologists who work with pedigree data, and describe successive iterations through approaches and prototypes for viewing genotyping errors in the context of a displayed pedigree. We describe how each approach performs with real pedigree genotype data and why eventually we deemed them unsuitable. Finally, a novel prototype visualisation for pedigrees, which we term the ‘sandwich view’, is detailed and we demonstrate how the approach effectively communicates errors in the pedigree context, supporting the biologist in the error identification task.

Nowadays, component-based development (CBD) has been widely used in software engineering. As a core activity in the CBD process, component retrieval has attracted extensive research attention. Despite the research efforts so far, one major problem still exists in component retrieval, that is, the mismatch between user's query and identified components. This paper proposes a semantics-base approach to tackle the mismatch issue in component retrieval process effectively. A review is first presented on the current state of relevant technologies involved in semantics-based component retrieval, and on the current development in this field. Component retrieval rating factors and their calculating method are proposed. The rating factors are used to illustrate and improve the precision of component retrieval in quantitatively.

We describe a case study of TaxVis, a multiple view system for examining relationships between sets of multiple classification trees. The system displays multiform views of the dataset, which in turn can either be a singular view of the larger forest object formed from the set of trees, or multiple views in themselves, using linking to show relationships between the separate trees. We describe the circumstances that led to the development of the multiple view aspects of the application and how it emerged that different views were indeed suited to different tasks.

We present a Directed Acyclic Graph visualisation designed to allow interaction with a set of multiple classification trees, specifically to find overlaps and differences between groups of trees and individual trees. The work is motivated by the need to find a representation for multiple trees that has the space-saving property of a general graph representation and the intuitive parent-child direction cues present in individual representation of trees. Using example taxonomic data sets, we describe augmentations to the common barycenter DAG layout method that reveal shared sets of child nodes between common parents in a clearer manner. Other interactions such as displaying the multiple ancestor paths of a node when it occurs in several trees, and revealing intersecting sibling sets within the context of a single DAG representation are also discussed.

Natural history museum collections contain a wealth of specimen level data that is now opening up for digital access. However, current interfaces to access and manipulate this data are standard text-based query mechanisms, giving no leeway for exploratory investigation of the collections. By adapting previous work on multiple taxonomies we allow visual comparison of related museum collections to discover areas of overlap, naming errors, and unique sections of a collection, indicating areas of specialisation for individual collections and the complementarities of the set formed by the collections as a whole.

OPAL (Online PArtner Lens) is an application designed to match project requirements with suitable teams and individuals, and as part of its matching process features an evaluation mechanism designed to elicit measures of trust between potential partners. We describe a matrix-style visualisation that displays these hierarchically structured assessments between sets of OPAL users to allow them to select potential partners. The main feature of the matrix visualisation is the ability for users to assess the context of a specific assessment as the visualisation not only reveals simple related statistics for the two users concerned, but also overlays summaries of related assessor and candidate evaluations as compact and ordered 'value bars' when the user examines information in the matrix. This enables the user to better decide whether a given assessment is in line with what would be expected from an assessor's and candidate's history, or whether it indicates a specifically localised interplay between the two users. Other features include a simple focus+context effect that can reveal the tree-like structure and details of assessments, and filtering assessments by their position in the matrix or by particular assessment attributes.

We introduce a number of refinements to the parallel coordinates visualisation metaphor for multidimensional data. Firstly, the traditional set of polylines are replaced with a collection of smooth curves across the attribute axes, allowing individual data elements to be traced under certain limitations; normally impossible due to the 'crossing problem'. Then the notion of spreading out points on axes with a few discrete values is introduced, which leads onto a simple focus+context technique when the user selects values on such axes.

Current techniques for visualising large-scale microarray data are unable to present temporal features without reducing the number of elements being displayed. This paper introduces a technique that overcomes this problem by combining a novel display technique, which operates over a continuous temporal subset of the time series, with direct manipulation of the parameters defining the subset.

This paper discusses issues and solutions for supporting multiple overlapping classifications in database systems. These classifications are commonly found in science, although they are often ignored in computing applications for scientific data, and inappropriate solutions adopted as their replacement. Known database models and classification techniques offer some degree of support for multiple overlapping classifications, but do not fully support the basic features we have identified as necessary: trees/graphs, traceability, semantics of classifications, independence of classification and data, and identity of classifications. The approach to the problem adopted by the Prometheus project, based on an extended object-oriented database model and the independence of classification schemes from classified data, is presented and discussed.

Visualisations have previously combined linking and brushing operations successfully. However, none (to our knowledge) have combined linking and focusing techniques. We describe the motivation behind developing such a technique, and show how our visualisation takes advantage of inherent links in a set of multiple overlapping hierarchies to build a mechanism that can produce linked focus+context effects between these multiple hierarchies with a single user action. Previously, our visualisation had utilised linking only, using colour to highlight correlations between distinct but overlapping hierarchies.

One of the tasks of plant taxonomy is the creation of classifications of organisms that allows the understanding of the evolutionary relationships between them. In this paper, we describe two different data models that have been designed to support two aspects of taxonomic work: the storage of the information and the visualisation of that information. We show that these two models are different because of their constraints and aims, and we compare their abilities using a number of typical tasks that users perform. We also show that, although different and able to perform different tasks, each of these models is well adapted to its purpose, and tight integration is difficult.

Model-based user interface development environments show promise for improving the productivity of user interface developers, and possibly for improving the quality of developed interfaces. While model-based techniques have previously been applied to the area of database interfaces, they have not been specifically targeted at the important area of object database applications. Such applications make use of models that are semantically richer than their relational counterparts in terms of both data structures and application functionality. In general, model-based techniques have not addressed how the information referenced in such applications is manifested within the described models, and is utilised within the generated interface itself. This lack of experience with such systems has led to many model-based projects providing minimal support for certain features that are essential to such data intensive applications, and has prevented object database interface developers in particular from benefiting from model-based techniques. This paper presents the Teallach model-based user interface development environment for object databases, describing the models it supports, the relationships between these models, the tool used to construct interfaces using the models and the generation of Java programs from the declarative models. Distinctive features of Teallach include comprehensive facilities for linking models, a flexible development method, an open architecture, and the generation of running applications based on the models constructed by designers. © 2001 Elsevier Science B.V. All rights reserved.

This short paper describes the presentation model used by the Teallach model-based user-interface development environment. Teallach's presentation model provides both abstract and concrete interactors, which are first-class objects that may be freely intermixed when building a user-interface. An example is provided showing this approach in use.

The visualisation of hierarchical information sets has been a staple of Information Visualisation since the field came into being in the early 1990's. However, at present, support for visualising the correlations between multiple, overlapping sets of hierarchical information has been lacking. This is despite the realisation that for certain tasks this information is as important as the information that forms the individual hierarchies. In response to this, we have produced two early visualisation prototypes, one based on a graph visualisation, and the other on a set-based metaphor, that endeavour to display such information in a readily perceived form to potential users. The science of botanical taxonomy is used as an example of a field where such a visualisation would be useful, and also as a resource for example information sets that the prototypes can act upon. Technical and perceptual issues involved in the design and implementation of both prototypes are discussed. Following this, informal user testing on both prototypes is described, which utilised user observation techniques to elicit qualitative feedback from the taxonomists. These findings are then used to emphasise the shortcomings and advantages of each prototype, and from these probable issues for future prototyping and development are drawn.

M.R. Pullen et al. (2000) have designed a new model of plant taxonomy (called Prometheus); it supports multiple overlapping classifications, and distinguishes the process of naming from classifying. The concepts identified in this taxonomic model necessitated the design of a new database model - the Prometheus Object-Oriented Model (POOM) - to represent and manipulate the data. POOM is an extended object-oriented model which emphasises relationships, thereby providing graph behaviour in an object-oriented database and providing an expressive means of defining relationships between objects. Additionally, the Object Query Language (OQL) is extended to the Prometheus Object-Oriented Language (POOL) in order to provide unified querying of object-oriented graph structures. This paper presents a taxonomic database system designed in terms of the concepts offered by POOM. Through examples we show how the representation of the semantics and processes of taxonomy, not possible using existing data models, can be supported. Example POOL queries highlight the need for the extended features for manipulating relationships, graph structures and complex objects such as are found in taxonomies.

This paper presents the work carried out in the Prometheus project and its motivation, taxonomy. Taxonomy presents challenges to common database systems. Because of its complexity and the necessary treatments applied to its data, common database model such as the relational, the object-oriented of even graph models are not able to support taxonomic applications fully. Our approach is the extension of a object-oriented database model with explicit relationships in order to support new features and thereby offer the necessary level of service for developing taxonomic applications.