MLVis 2018

Machine Learning Methods in Visualisation for Big Data 2018
Tutorial co-located with EuroVis 2018, June 4, 2018, Brno, Czech Republic

MLVis 2018

Machine Learning Methods in Visualisation for Big Data 2018

Tutorial co-located with EuroVis 2018, June 4, 2018, Brno, Czech Republic

In order to handle big data challenges, machine learning techniques can be advantageous in simplifying and summarising large data sets for visualisation. Machine learning provides methods that allow the summarisation of very large data sets whereas visualisation leverages the human visual system to help find unanticipated patterns. In this tutorial, we cover machine learning methods relevant to the area of visualisation. In addition to an exploration of the applicability, strengths, and weaknesses of such approaches, we provide links to available software tools that can help provide solutions to machine learning problems.

Main Topics of the Tutorial

  • Dimensionality Reduction for Visualization
  • Sampling Methods and Graph Visualisation
  • Evaluating Visualisation Techniques

In addition to the tutorial topics the event features a paper presentation, a panel discussion, and a Data Lab: Bring Your Own Data session.

Call for Papers

Machine Learning Methods in Visualisation will be held as part of EuroVis 2018 in Brno, Czech Republic. After two years of tutorials, the third edition of this co-located event will have a new format to become part-tutorial and part-workshop so as to increase the interaction between researchers. This year, part of the programme will consist of short papers from the machine learning and visualisation communities on how the two technologies can be used together to provide greater insight to end users.

MLVis 2018 solicits short papers of 4 pages excluding references (5 pages in total) on the topic of machine learning methods in visualisation. These should be formatted using the EuroVis 2018 style.

Please use the attached style files for LaTeX to format your submission.

The short papers will be published on the EuroVis memory stick and as part of the EG Digital Library. Papers are intended to present works in progress and will be presented during the workshop as a 15 minute presentation (15min + 5min questions). Papers must have both a machine learning and visualisation component or must influence both fields in some way.

The workshop chairs this year plan to review submissions in a single round review cycle. All submissions should be made using the Precision Conference System (PCS).

In the Precision Conference System, please choose EuroVis as the society and MLVis 2018 as the conference when creating your submission.

Accepted papers will be presented during the workshop at EuroVis 2018

Deadlines:

  • Submission: March 5, 2018 (extended deadline)
  • Notification: April 3, 2018
  • Camera Ready Version: April 20, 2018

Programme

  • 9:05-9:15 Introduction
  • 9:15-9:45 Dimensionality Reduction for Visualization (Presenter: Jaakko Peltonen)
    • Component analysis methods
    • Self-organizing maps
    • Multidimensional Scaling (MDS) and its variants
    • Methods that preserve similarities or neighbourhood relationships

    Slides in PDF format

  • 9:45-10:15 Sampling Methods and Graph Visualisation (Presenter: Daniel Archambault)
    • Influence of sampling on perception of graph structure
    • Sampling across time on dynamic graphs.

    Slides in PDF format

  • 10:15-10:40 Paper presentation: “Panning for Insight: Amplifying Insight through Tight Integration of Machine Learning, Data Mining, and Visualization”, Benjamin Karer, Inga Scheler and Hans Hagen, TU Kaiserslautern, Germany
    Slides in PDF format (note: full paper available as part of the Eurovis 2018 co-located event proceedings)

  • 10:40-11:10 Coffee Break
  • 11:10-11:45 Evaluating Visualisation Techniques (Presenter: Ian Nabney)
    • Why is evaluation important yet difficult?
    • User-based evaluation: perceptual evaluation, study design
    • Metric-based evaluation: model-based metrics, unsupervised learning metrics, task-based metrics

    Slides in PDF format

  • 11:45-12:15 Panel Discussion
    Tentative topics:

    • Future fruitful lines of research
    • Obstacles/challenges to increased combination of ML and vis
    • How to evaluate ML&vis at different levels, and convince others of the evaluation?
    • How do we relate do data science?
    • Topics for the next CFP
  • 12:15-12:45 Data Lab: Bring Your Own Data
  • 12:45-12:50 Closing

Registration

Registration to the workshop will be handled as part of the general registration to Eurovis 2018.

Organisation

Organisers:

  • Ian Nabney is Professor and Head of SCEEM School of Engineering at University of Bristol. He received his BA in Mathematics from Oxford University and a PhD in Mathematics from Cambridge University. He has over 20 years’ experience in machine learning research, has published more than 80 papers (1900 citations), and is the system architect for the Netlab pattern analysis toolbox, which has been downloaded more than 40,000 times since 1999 (the accompanying book has been through three reprints), and the Data Visualisation and Modelling System (DVMS) which integrates data projection and information visualisation techniques to provide a rich interactive environment for data exploration and visual analytics. DVMS will be used for the demonstrations of generative models. He has won grants worth more than 3M GBP from EPSRC, the EU, TSB, and industry and has supervised 11 PhD students to completion. He is the Chair of the Natural Computing Applications Forum, a principal mechanism in the UK for exchange of ideas between academics and industry on natural computing technology and practical applications.
  • Jaakko Peltonen is a professor of statistics (data analysis) at the Faculty of Natural Sciences, University of Tampere where he leads the Statistical Machine Learning and Exploratory Data Analysis research group; he is also currently visiting professor at the Department of Computer Science, Aalto University. He received his D.Sc. from Helsinki University of Technology in 2004. He is an editorial board member of Neural Networks, associate editor of Neural Processing Letters, editorial board member of Heliyon, and executive committee member of the European Neural Network Society. He has served in organising committees of eight international conferences and in program committees of 44 international conferences/workshops, and has referee duties for numerous international journals and conferences. He is an expert in statistical machine learning methods for exploratory data analysis, visualisation of data, and learning from multiple sources.
  • Daniel Archambault received his PhD in Computer Science from the University of British Columbia, Canada in 2008. He is currently a Senior Lecturer of Computer Science at Swansea University in the United Kingdom. During his post-doctoral studies at University College Dublin, he applied his expertise in information visualisation to help visualise the results of machine learning approaches, particularly in the area of social media visualisation. This work inspired him to co-chair the AAAI ICWSM Workshop on Social Media Visualisation (SocMedVis 2012 and 2013). His other areas expertise primarily lie in graph visualisation and drawing as well as perceptual factors in information visualisation.