Objectives and Work Packages

 

NeEDS Objectives

This network addresses the urgent need for an integrated modelling and computing environment that facilitates data processing, data analysis and data communication (in the form of visualization and human-computer interaction) to aid decision making.

The main scientific and technological objectives of NeEDS are to develop innovative mathematical optimization models and high performance algorithms

  • for novel applications involving Network Science;
  • to ensure Interpretability, fulfilling the right-to-explanation in algorithmic decision making required by the EU as of 2018, but also required when nonexperts are to interact with data analysis tools;
  • to deal with the challenges posed by Complex Data such as time-evolving data, spatial data, and process data;
  • to Extract Knowledge from data by jointly addressing data processing and data analysis.

NeEDS bridges the disciplines of Computer Science, Business Analytics, Mathematical Optimization and Statistics to achieve a breakthrough that requires an interdisciplinary approach, namely, the development of computational methods that are easy-to-interpret and easy-to-interact with, that run under strict time regulations, and that can cope with uncertainty in data fluctuation.


NeEDS Research Work Packages

Work Package 1. Developing innovative tools to tackle Network data

This work package is led by the team at Katholieke Universiteit Leuven (KUL). The group has made landmark contributions in the area of Business Analytics, such as in Credit Risk Management. With two Research & Innovation Projects, Work Package 1 addresses cutting-edge issues in Network Science, motivated by the pressing needs in the banking and the insurance industry.

So far, no videos of WP1 secondments are available but we continuously update the list.


Work Package 2. Cutting-edge modelling to enhance Interpretability

This work package is led by the team at Copenhagen Business School (CBS). This is a group with a long-standing expertise in the field of Mathematical Optimization, and a key contributor to the area of Analytics and Big Data. Work package 2 advances the state-of-the-art in the well-established field of interpretable Data Science tools, improving performance and developing novel measures of interpretability.

For more information on what our Early Stage Researchers (ESRs) have been working on in WP2, click on the names below and take a look at their video presentations:

Geographica     Marcela Galvis Restrepo (Copenhagen Business School)

Duke University     Jeanette Walldorf (Copenhagen Business School)

Danmarks Statistik     Thomas Ian Ashley (Universidad de Sevilla)

Danmarks Statistik     Cristina Molero del Río (Universidad de Sevilla)

Universidad de Sevilla     Jonas Klingwort (Centraal Bureau voor de Statistiek)


Work Package 3. Addressing the challenges of Complex data arising in Industry

This work package is led by the team at University of Oxford (UOXF). This group has a world-class track record in carrying out research in Visual Analytics, is a key contributor to the development of this burgeoning area, and has an invaluable expertise in knowledge transfer activities. Work Package 3 addresses the problem of how to analyze complex data, namely event and process data that vary over time, that abound at our industrial participants.

For more information on what our Early Stage Researchers (ESRs) have been working on in WP3, click on the names below and take a look at their video presentations:

Centraal Bureau Voor de Statistiek     Yu Zhang (UOXF)


Work Package 4. Innovative Extraction of knowledge by jointly addressing data processing and data analysis

This work package is led by the team at Universidad de Sevilla. This is a group with a long-standing expertise in the field of Mathematical Optimization, and a key contributor to the area of Mathematical Optimization and Supervised as well as Unsupervised Learning. Work Package 4 extracts knowledge by jointly addressing data processing and data analysis, a nascent and challenging area of research with a huge potential in terms of improved performance of Data Science tools. The knowledge extracted will be used to build scenarios to deal with data uncertainty, which is at the heart of the problems faced by our industrial participants.

For more information on what our Early Stage Researchers (ESRs) have been working on in WP4, click on the names below and take a look at their video presentations:

Danmarks Statistik     María Remedios Sillero Denamiel (Universidad de Sevilla)

Danmarks Statistik     Sandra Benítez-Peña (Universidad de Sevilla)