Skip to main content
Go to the home page of the European Commission (opens in new window)
English en
CORDIS - EU research results
CORDIS

The design and evaluation of modern fully dynamic data structures

Project description

New fully dynamic data structures to support data mining and network analysis

Every computer programme uses data structures. Since data sets today frequently are large in size, dynamic and can change continuously, it is not always possible to re-process all the data. The EU-funded MoDynStruct project will develop techniques to query information in dynamic databases, providing an important solution for data mining and network analysis. Moreover, the techniques under development will ensure that the private information contained in the data sets is protected. The project will focus on problems of large practical relevance, such as subgraph detection and various clustering variants in general metric spaces, for which no fully dynamic data structures with small asymptotic running time currently exist.

Objective

Many real-world data sets change continuously, but their enormous size prohibits frequent re-processing of the whole data. Thus, there is an urgent need for efficient, fully dynamic data structures that maintain properties of the data set while supporting fast insertions and deletions. This is especially important for problems in data mining and network analysis, where a data structure often needs to fulfill new additional constraints that are not supported by classic data structures: (1) It should only use sublinear space, even if this leads to some small error in the answers. (2) As data sets frequently contain private information which needs to be protected, it should reveal nothing about individual data points, which is often modeled through differential privacy. Our ambitious goal is to design such groundbreaking new fully dynamic data structures for central problems on graphs and point sets.

Specifically, we will focus on problems with large practical relevance such as subgraph detection, k-core decomposition, and balanced graph partitioning as well as various clustering variants in general metric spaces. For these problems no fully dynamic data structures with small asymptotic running time are known and they have not even been studied in the small-space or differentially-private regime. However, using recent advanced in algorithms research it is now the right time to develop novel techniques to solve these challenging questions.

Thus, the goal of this project is to design algorithms for highly-relevant problems as well as advancing the field of data structures in general by moving it from a narrow focus on asymptotic complexity to a broader set of modern requirements with the goal of bridging the gap that currently exists between theory and practice. As data structures are used by every computer program the impact of this work will be far-reaching. With over 30 years of experience in algorithms research the PI is in the unique position to do so.

Fields of science (EuroSciVoc)

Keywords

Programme(s)

Topic(s)

Funding Scheme

ERC-ADG - Advanced Grant

See all projects funded under this funding scheme

Call for proposal

(opens in new window) ERC-2020-ADG

See all projects funded under this call

Host institution

INSTITUTE OF SCIENCE AND TECHNOLOGY AUSTRIA
Net EU contribution
€ 2 389 169,75
Address
Am Campus 1
3400 Klosterneuburg
Austria
Region
Ostösterreich Niederösterreich Wiener Umland/Nordteil
Activity type
Higher or Secondary Education Establishments
Links
Total cost
€ 2 389 169,75

Beneficiaries (2)