Program

Overview

The workshop will be divided on four sections distributed in four sessions according the AVI timetable. After a short introductory presentation and the road mapping methodology presentation, every research group will be presenting their paper within a 20 minutes’ timeframe. To define the demand of new reference models and infrastructures for Big Data applications as well as to define the basis for the road map, every presentation should focus on three questions; Which application domain are addressed, which problem should be solved, and which Big Data technologies are used? After the presentations of the participants, the organizers will present a “Reference Model for Advanced Visual Interfaces Supporting Big Data Analysis”, that will be the framework for the road map development. The third section contains a sub group working session, where the participants are elaborating how their research can be included in the introduced reference model. Therefore, the organizers provide each participant a power point template that has to be prepared and presented within the fourth section. The last section contains the road map development that will be archived as a result of the sub group presentation in a cooperatively discussion. The wrap up and the definition of the next steps with the corresponding timeline completes the Workshop.

The initial main outcome will be a common research agenda, which will be a good starting point for possible proposals of joint research activities. Another initial outcome could be an article in an international journal describing the common research road map. Therefore, every participant gets the opportunity to combine their initial paper with the results of the sub group working sessions, to submit an extended version. Revised and accepted papers will be published as a post-proceeding within the Springer Lecture Notes in Computer Science (LNCS) series (www.springer.com/lncs).


Schedule

Session 1 (09:00 – 11:00)

Sec 1: 09:00 – 09:15 Short introductory presentations

Sec 1: 09:15 – 09:30 Introduction of road-mapping methodology and discussion

Sec 2: 09:30 – 09:50 Presentation: Engineering Study of Tidal Stream Renewable Energy Generation and Visualization: Issues of Process Modeling and Implementation

Sec 2: 09:50 – 10:10 Presentation: Cost effective Visualization of Research Data for Cognitive Development Using Mobile Augmented Reality

Sec 2: 10:10 – 10:30 Presentation: Rapidly Visualizing NGS Cancer Data Sets with Cloud Computing

Sec 2: 10:30 – 10:50 Presentation: SenseCare: Towards an Experimental Platform for Home-Based, Visualisation of Emotional States of People with Dementia

Coffee break

Session 2 (11:30 – 13:00)

Sec 2: 11:30 – 11:50 Presentation: Towards Interactive Visualization of Results from Domain Specific Text Analytics

Sec 2: 11:50 – 12:10 Presentation: Modern Technologies to Synchronize Data Sources and Information Visualizations

Sec 2: 12:10 – 12:30 Presentation: Visual analytics and mining over Big Data. Discussing some issues and challenges, and presenting a few experiences

Sec 2: 12:30 – 12:50 Presentation: A Meta-Design Approach to Support Information Access and Manipulation in Virtual Research Environments

Lunch

Session 3 (14:30 – 16:00)

Sec 2: 14:30 – 14:50 Presentation: Towards a Reference Model for Advanced Visual Interfaces Supporting Big Data Analysis

Sec 2: 14:50 – 15:20 Discussion

Sec 3: 15:20 – 16:00 Preparation of sub-group working session

Afternoon Coffee

Session 4 (16:30 – 18:00)

Sec 3: 16:00 – 16:45 Sub-group working session

Sec 4: 16:45 – 17:45 Presentation of sub-group working session results, discussion, and road map development

Sec 4: 17:45 – 18:00 Wrap Up – Next steps and timeline


Accepted Papers

Towards Interactive Visualization of Results from Domain Specific Text Analytics

Tobias Swoboda1, Christian Nawroth1, and Michael Kaufmann2.
1University of Hagen, Faculty of Mathematics and Computer Science;
2Lucerne University of Applied Sciences and Arts, School of Engineering and Architecture.

In Big Data analytics, visualization and access are central for the creation of knowledge and value from data. Interactive visualizations of analysis of structured data are commonplace. In this paper, information visualization and interaction for the text analysis is addressed. The paper motivates this issue from a data usage standpoint, presents a survey of approaches in the area of interactive visualization of text analytics, and presents our proposal of a specific solution design for visual interaction with results from a combination of named entity recognition (NER) and text categorization (TC). This matrix-based model illustrates abstract views on complex relationships between abstract entities and is exemplary for any combination of feature extraction and TC. The aim of our example is to support feature extraction and TC researchers in distributed virtual research environments by providing intuitive visual Interfaces.

A Meta-Design Approach to Support Information Access and Manipulation in Virtual Research Environments

Carmelo Ardito1, Paolo Buono1,  M. Francesca Costabile1,  Giuseppe Desolda1,  and Maristella Matera2.
1Università degli Studi di Bari Aldo Moro;
2DEIB, Politecnico di Milano.

Virtual Research Environments (VREs). i.e., distributed and dynamic environments that foster the collaboration of researchers from different disciplines by supporting the accomplishment of complex research tasks, lack efficient and effective user interfaces addressing user diversity. In this paper we illustrate the meta-design approach adopted in the last years to design user interfaces adequate for different user communities: This approach is in particular suitable to address the variety arising in the cultural background of users, their reasoning strategies, the way they carry out their tasks in their daily practices, the languages and notations they are familiar with. We also describe a mashup platform, developed on the basis of a meta-design model, that enables end users to extract contents from heterogeneous sources and manipulate such content in their personal interactive environments, thus creating new content that can be shared among people collaborating to a certain task.

Modern Technologies to Synchronize Data Sources and Information Visualizations

Christian Danowski1, Marco X. Bornschlegl1, Benno Schmidt2, and Matthias L. Hemmje1.
1University of Hagen, Faculty of Mathematics and Computer Science;
2Bochum University of Applied Sciences, Department of Geodesy.

There are many systems that visualize abstract data in the context of information visualization (IVIS). However, most systems create unidirectional visualizations that represent a static product of the source data. Any changes to the content of the visualization – if that is possible at all – has no effect on the source data. This paper recommends an approach to use modern Web technologies and state-of-the-art system architectures to develop a bidirectional IVIS system where the client-user is able to alter data properties within three-dimensional visualizations (using the ISO-specified X3D format of the Web3D Consortium) that in consequence are automatically applied to the server-side data sources. Furthermore relevant data changes on the server side are published to all registered clients to maintain a permanent consistent state. Particularly, a mediator-wrapper architecture is used in order to semantically integrate heterogeneous data sources in Big Data scenarios.

Towards a Reference Model for Advanced Visual Interfaces Supporting Big Data Analysis

Marco X. Bornschlegl1, Kevin Berwind1, Michael Kaufmann2, Felix C. Engel1, Paul Walsh3, and Matthias L. Hemmje1.
1University of Hagen, Faculty of Mathematics and Computer Science;
2Lucerne University of Applied Sciences and Arts, School of Engineering and Architecture;
3Cork Institute of Technology, CIT Informatics.

This paper introduces an approach to develop an up-to-date reference model that can support advanced visual user interfaces for distributed Big Data analysis in virtual labs to be used in e-Science, industrial research, and data science education. The paper introduces and motivates the current situation in this application area as a basis for a corresponding problem statement that is utilized to derived goals and objectives of the approach. Furthermore, the relevant state-ofthe- art is revisited and remaining challenges are identifed. An exemplar set of use cases, corresponding user stereotypes as well as a conceptual design model to address these challenges is introduced. The scenario and user stereotypes have been an expert roundtable. A corresponding architectural system model is suggested as a conceptual reference architecture to support proof-of-concept implementations as well as to support interoperability in distributed infrastructures. Conclusions and an outlook on future work complete the paper.

Visual analytics and mining over Big Data. Discussing some issues and challenges, and presenting a few experiences

Marco Angelini, Tiziana Catarci, Massimo Mecella, and Giuseppe Santucci.
Sapienza Universita di Roma, Dipartimento di Ingegneria Informatica Automatica e Gestionale

In this short position paper, we present a few concrete experiences of Visual Analytics over big data; as our experiences have been gained on the application domains of cyber security and OSINT, which are very relevant and crucial domains targets of possible VREs, we also discuss and propose an high-level reference architecture and pipeline for a BigData service in VREs dealing with such aspects, in which the VA part is crucial in order to provide effectiveness tousers.

Cost effective Visualization of Research Data for Cognitive Development Using Mobile Augmented Reality

C. Onime1, and J. O. Uhomoibi2.
1The Abdus Salam International Centre for Theoretical Physics (ICTP);
2Ulster University, Faculty of Computing and Engineering;

In many Fields of scientific research, the numerical output of research work require proper interpretation in relation to real world situations. Graphical visualization is often used to ensure better comprehension of data (research outputs) by researchers, learners and other stakeholders. However, In the modern era, large scale experimentation as well as computer based simulations are generating massive amounts of numeric data that are almost impossible to visualize using traditional plots and graphs as they are limited in both dimensions and scale. Video has gained increasing popularity for presenting data due to its ability to convey motion and time. While, such video presentations are undoubtedly useful, they provide limited contributions to cognitive development. In this short paper, a cost effective use of mobile augmented reality in the visualization of scientific research data highlighting two use-cases that show the Three Dimensional (3D) semi-immersive and interactive environment in both educational and non-educational contexts, will be examined.

Engineering Study of Tidal Stream Renewable Energy Generation and Visualization: Issues of Process Modeling and Implementation

J. Harrison, and J. O. Uhomoibi.
Ulster University, Faculty of Computing and Engineering;

Tidal stream energy has the potential to make a significant contribution to energy mix in the future. Accurate modeling and visualization of both tidal resource and array layout enhances understanding of in-stream tidal behavior leading to improvements in site identification and optimal positioning of individual turbines. A realistic representation of blade loading conditions will aid designers and manufacturers in creating more robust devices and improve survivability. The main barriers to large scale deployments of tidal arrays are the costs associated with manufacturing, installation and maintenance. Therefore, presently tidal energy is not competitive on cost with more established renewable technologies. The current position paper investigates and reports on resource modeling, site selection, selecting optimal array configurations and the design and manufacture of devices for tidal stream renewable energy generation. This is aimed at developing models to reliably simulate real conditions, enhance understanding of tidal processes, flow regimes and device survivability issues.

SenseCare: Towards an Experimental Platform for Home-Based, Visualisation of Emotional States of People with Dementia

Felix C. Engel1, Raymond Bond2, Alfie Keary3, Maurice Mulvena2, Paul Walsh3, Zheng Huiru2, Ulrich Kowohl4, and Matthias L. Hemmje1.
1Research Institute for Telecommunication and Cooperation;
2Ulster University, Faculty of Computing and Engineering;
3NSilico Life Science;
3University of Hagen, Faculty of Mathematics and Computer Science.

In this paper, the support and care of people with dementia in their own homes is explored. The paper sets out a framework-, to gather and analyze data in the homes of people with dementia. Basic requirements are explored towards development of the SenseCare platform stemming from application scenarios in which various data streams are created, processed, analyzed, visualized and stored for ad-hoc or later reuse. The framework will be realized as a platform based on open ICT standards, implemented within the EC co funded SenseCare project. In this publication the components of the SenseCare platform are described, including the visualization of data.

Rapidly Visualizing NGS Cancer Data Sets with Cloud Computing

Paul Walsh1, Brendon Lawlor1, Brian Kelly1, Michael Bekaert1, Timmy Manning1, Timm Heuss1, Xiangwu Lu1, Roy Sleator1, and Markus Leopold2.
1NSilico Life Science;
2Darmstadt University of Applied Sciences.

With the advent of NGS technology, clinical data sets now contain enormous amounts of valuable genomic information related to a wide range of diseases such as cancer. This data needs to be analyzed, managed, stored, visualized and integrated in order to be clinically useful. However, many clinicians and researchers, who need to interpret these data sets, are non-specialists in the IT domain and so need systems that are effective and easy to use. Herein, we present an overview of a novel cloud computing based NGS research management software system which has simplicity, scalability, speed and reproducibility at its core. The efficacy of the system is demonstrated by showing how the system enables rapid visualization of big data cancer sets.  We present results from a bio-informatics pipeline run by Simplicity™ in Sage-Care, an EU funded cancer research project, for comprehensive genome mapping analysis and visualization.