giCentre at IEEE Vis 2014

This year IEEE VIS was held in Europe (Paris) for the first time. giCentre researchers made contributions throughout the week.

VIS kicked off with a series workshops. Rafael Henkin presented an entry to the VAST Challenge that was given an Honourable Mention and Natalie and Gennady Andrienko presented an award winning entry using their Semantic Spaces methodology. Roger Beecham delivered a paper on Map Line-ups at the  DECISIVe workshop, Cagatay Turkay presented at a workshop on Visualization for Predictive Analytics and Jason Dykes co-lead a workshop on the Future of Vis.

During the conference itself, giCentre researchers were involved in four paper presentations:   Attribute Signatures, an approach to analysing multivariate data; Stenomaps, a new technique for simplified representations of spatial data; Perceptually Uniform Motion Space, a study of factors that affect motion perception; and a Multi-channel Approach to Data Visualization, which reflects on four years of collaborative design and analysis with colleagues at TfL. In addition, Sarah Goodwin presented a well-received poster on Multivariate Correlation and having been papers chair in 2012 and 2013, Jason chaired this year's IEEE Information Visualization conference.

As well as enjoying some Parisian gastronomy, the conference culminated with a bike ride around Paris and its hinterlands. 

giCentre at GIScience

giScience.jpg

giCentre researchers attended the Eighth International Conference on Geographic Information Science at the Vienna University of Technology. This bi-annual forum is the premier venue for research in the discipline.

An ICA Commission on GeoVisualization Workshop on GeoVisual Analytics had well received contributions from PhD candidate Sarah Goodwin and Dr Cagatay Turkay whose approaches allow analysts to interactively assess the effects of scale as they explore multiple attributes of geographic locations.

The meeting involved a touching tribute to giCentre friend and colleague Professor Peter Fisher who passed away suddenly in May. Pete received a minute's applause. Jason Dykes also paid tribute to Pete, who supervised Jason's PhD, in a keynote lecture in which he described giCentre approaches to creative cartography that combine Information Visualization and more traditional cartography. You can find links to the various resources used and mentioned in an accompanying blog posting.

Research Assistant in Visual Analytics

We are looking to hire a Research Assistant in Visual Analytics / Machine Learning as part of the FareViz project at City University London. The project involves developing new ways of visualizing and analysing ticket pricing of rail journeys in the UK. Its aim is to reveal structure and anomalies in the complexities of UK rail pricing to support the general public, transport analysis/policy and ticket vendors.

The 15 month full-time post will involve developing new visual analytic software combining interactive visual exploration with machine learning based approaches to analysis. The project is funded by the ESRC / Technology Strategy Board in partnership with Transport API, RailEasy and DigitalMR. The deadline for application is 7th September 2014.

Further details and online application.

Research Officer post: Big Data in Survey Research

We are looking to hire a Research Officer to use Big Data in survey research undertaken in collaboration with the Centre for Comparative Social Surveys (CCSS) at City University London. The post will be based in CCSS in the School of Social Sciences and will involve working closely with the giCentre. CCSS undertake the European Social Survey (ESS), which measures, monitors and interprets changing social attitudes in Europe and develops methods for cross-national research. The giCentre develop analytical visual interfaces to make sense of data, often with a spatial emphasis.

The post is central to an 18-month ESRC-funded research project - Using multi-level multi-source auxiliary data to investigate non-response bias in UK general social surveys, through which CCSS and giCentre will explore the extent to which the predictive power of Big Data can be used to address poor survey response rates. The Research Officer will be required to work on all parts of the project, including scoping suitable auxiliary data sources, conducting non-response analysis, producing non-response weights, conducting geographically weighted spatial analysis and using data visualisation to derive and present results. 

Further details and online application.   

Research Associate Position in Visual Analytics

The giCentre is recruiting a Research Associate to work on the EU funded Project VALCRI.

The project will use novel modes of interaction to develop visual analytics tools that support criminal intelligence analysts in making sense of the masses of data that they use to generate evidence. We will be designing and building visualization tools to provide access to these data, detect patterns in them and support analysts in using them to develop arguments, threads of evidence and stories.

VALCRI is a collaboration between 18 partners across Europe, with coordination at Middlesex University.  This is a full-time post with a fixed-term until December 2017. Further details and application.

PacificVis best paper award

Joint work between TU Eindhoven and the giCentre has won Best Paper at IEEE PacificVis in Yokohama, Japan. The paper presents a new way of schematising maps using curved edges that preserve the relative area of features.

The technique offers new design possibilities for geovisualization  that improve upon traditional automated simplification routines.

PhD scholarships available

phd.jpg

The Department of Computer Science is offering three Doctoral Scholarships each valued at around £50,000 over three years. This provides an ideal opportunity for outstanding researchers to join the giCentre team. Applications must be completed by 14th March 2014.

The deadline for 2014 scholarships has now passed, but we still welcome enquires from anyone wishing to study at the giCentre.