Research projects

A BETTER project for exploiting Big Data in Earth Observation

The main objective of BETTER is to implement an EO Big Data intermediate service layer devoted to harnessing the potential of the Copernicus and Sentinel European EO data directly from the needs of the users.The value will be demonstrated through a series of use-cases, which will be identified following a number of workshops and hackathons with potential users of this free and valuable resource. Big Data architectures, based on the BigDataEurope platform, will be customised for each use-case.

Project duration

November 2017 – October 2020

Link to the project page

REACH – Responsive Engagement of the Elderly Promoting Activity and Customized Healthcare

The project develops modular, ethically acceptable modular, personalized medical and ethical acceptable solutions integrated in and around buildings (clinical environments, rehabilitation settings, care homes, and home care), which allow an intelligent prediction (considering both personal medical history as well as real-time gathered data from a series of embedded sensors) about the health status of people/patients. Based on forecast and analytic algorithms REACH will allow to provide novel, personalized interventions (customized services, products, and equipment for mobilization and rehabilitation, physical activity, training, food and nutrition, mobility, and patient motivation). Our focus at IAIS are technologies for big data analytics, monitoring and decision support.

Project duration:

February 2016 – January 2010

SoBigData Research Infrastructure

SoBigData will create an ecosystem and research infrastructure for social mining in big data. It will enable easy comparison, re-use and integration of social big data, analytical tools and services into new research. We will contribute to the infrastructure which provides access to tools and services for social mining in data big repositories.

Project duration

September 2015 – August 2019

Link to the project page

datAcron – Big Data Analytics for Time Critical Mobility Forecasting

datACRON is about early detection of threat and abnormal activities in very large fleets spread across large geographical areas in sea and air. It investigates novel methods for real-time detection and prediction of trajectories and important events related to moving entities.

We focus on advanced visual analytics methods for multiple heterogeneous, voluminous, fluctuating, and noisy data streams from moving entities. Correlating them with archived data expressing entities’ characteristics and geographical information should expose mobility patterns, planned routes and intentions in a timely manner. 

Project duration

January 2016 – December 2018

Link to the project page

VaVel – Variety, Veracity, Value: Handling the Multiplicity of Urban Sensors

VaVeL addresses the most critical inefficiencies of current (big) data management and stream frameworks to cope with emerging urban sensor data thus making European urban data more accessible and easy to use for European industries.

We contribute highly scalable interactive visualization methods to support explorative data analysis and predictive modeling.

Project duration:

December 2015 – November 2018

Link to the project page

CITYCoP – Citizen Interaction Technologies Yield Community Policing

Theories underlying community policing received new impetus with the recent advent of smartphones and social media and especially user-generated content where citizens engage in closer interaction with their local community and law enforcement agency. CITYCoP goes on to develop a solution including a new smartphone app and on-line portal which are capable of being deployed in any European city while still retaining “local flavour” and diversity. These ICT solutions will also be designed from scratch to be fully compliant with strict privacy and data protection laws. Fraunhofer IAIS is contributing big data textmining technologies for social media monitoring.

Project duration:

June 2016 – May 2018

Link to the project page

GRACeFUL – Global Systems Rapid Assessment Tools through Constraint Functional Languages

The making of policies coping with global challenges like climate change and global financial crises is a process that necessarily involves the participation of stakeholders with very diverse backgrounds. A useful instrument to assess proposed solutions are simulators. Here the project investigates the benefits of functional, constraint-based languages. We contribute tools for visual analytics that support such languages. In addition we develop interactive visual interfaces for a rapid assessment tool, which can take into account the questions, perspectives and capabilities of different user groups.

Project duration:

February 2015 – January 2018

Link to the project page

EDSA – European Data Science Academy

In 2012 the Harvard Business Review declared “data scientist” to be the »s sexiest job of the 21st century«.  Ever since diverse studies observe that the demand for such experts for advanced data analytics and big data will not be satisfied for years to come.

The EDSA project wants to establish a »European Data Science Academy«. For preparation, it investigates the demand for data scientist across sectors and member states, designs corresponding curricula, and produces and evaluates learning materials. We contribute our data scientist training program and selected materials.

Project duration:

February 2015 – January 2016

Link to the project page

DART – Data-Driven Airport Trajectory Prediction Research

Air traffic management has reached the limits of predictability, efficiency and cost effectiveness. Different initiatives promote a new paradigm where trajectories are planned, negotiated, executed and amended. This requires a high-fidelity prediction of aircraft trajectories which takes into account the complexity of the whole network. In this project we investigate methods of data science and complexity science for such predictions.

Project duration:

January 2016 – December 2017

Smart Energy Hub

Instead of few big energy producers and many consumers the future will see a number of small and big energy producers and storages. This project faces the challenges of creating an intelligent network that controls energy consumption and storage automatically based on predictions. A pilot will be created for Stuttgart airport. Our contribution is a study on in-memory systems which shall guide the selection of a suitable bi data architecture.

Project duration:

January 2015 – December 2017

Link to the project page

Big Data Europe

This project creates the foundations for European companies to exploit big data and knowledge assets, which are semantically interoperable, available in different languages and under different licenses, for new products and services. We create a catalog of technical components with their features, functions and dependencies. We try to close gaps which require further research and development, such as a tighter integration of data mining and visual analytics.

Project duration:

January 2015 – December 2017

Link to the project page

iPRODICT – The Intelligent Factory

iPRODICT develops an intelligent approach to semi-automate the adaptation and improvement of business processes. Historic process information as well as real-time sensor data are used to predict the optimal process execution. Big Data analytics allows to adapt the process to the current context. An application scenario in the steel processing industry highlights the impacts of the approach in the areas of Industry 4.0 and factories of the future.

Project duration: 

September 2014 – August 2017

Link to the project page(german)

SAKE – Semantic Analysis of Complex Events

SAKE develops a modular framework for processing event data. A first application area is concerned with the timely recognition and prediction of operating flaws in production systems. A second application area deals with monitoring and detection of failures in IT networks. We will apply and evaluate semantic web standards for storing data suitable for machine learning.

Project duration: 

December 2014 – November 2017

Link to the project Page

FERARI – Flexible Event Processing for Big Data Architectures

FERARI paves the way for efficient and timely processing of Big Data. The systematic approach will enable leveraging recent advances in in-situ processing algorithms, which perform much of the processing at the source where the data is generated. By diminishing the need for large centralized infrastructures, huge data transfers, and the respective necessary energy, in-situ processing lowers the cost of Big Data stream processing systems by orders of magnitude. Similarly, huge acceleration is obtained in performing real-time knowledge extraction and monitoring. Applications include the analysis of M2M data and streaming fraud detection in telecommunications.

Project duration:

February 2014 – January 2017

Link to the project page

CAP4Access – Collective Awareness Platforms for Improving Accessibility in European Cities & Regions

The large group of persons with restricted mobility includes wheelchair users, elderly people with walking frames and parents with pushchairs. The web portal Wheelmap.org informs such persons about the accessibility of places. In CAP4Access we tap new sources with data about the character of pavements to improve a routing which can be adapted to user profiles corresponding to different physical capabilities. Moreover, we support awareness-raising  and community actions with web-based visualizations.

Project duration:

January 2014 – December 2016

Link to the project page

INSIGHT – Intelligent Synthesis and real-time Response using massive Streaming of heterogeneous Data

In this project we collaborate with the German Bundesamt für Bevölkerungsschutz und Katastrophenhilfe (BBK) to develop a prototype for fast situational assessmen. In near real-time various big data sources shall be monitored, such as mobile telephony, sensors, social media and news streams. Tools for the detection and visualization of relevant complex events and trends related to catastrophies shall facilitate the situational assessment at BBK.

Project duration:

September 2012 – August 2015

Link to the project page

EURECA

The goal of  Eureca was to enable seamless, secure, scalable and consistent linkage of healthcare information residing in electronic health recordsystems with information in clinical research information systems, such as clinical trials. We contributed to machine learning and data mining for medicine healthcare, with a particular focus on the mining of medical texts and personalized medical recommender systems.

Project duration:

February 2011 – July 2015

Link to the project page

p-Medicine

p-medicine aimed at developing new tools, IT infrastructure and VPH models to accelerate personalized medicine for the benefit of the patient. We contributed clinical decision support, big data analytics in medicine, and privacy-preserving data mining technologies.

Project duration:

February 2011 – July 2015

Link to the project page

MODAP - Mobility, Data Mining and Privacy

Mobile devices like smartphones and tablets have become wide-spread and produce an unbelievably large amount of mobility data. Since 2005 we have investigated application scenarios for using such data, first in the  project GeoPKDD, then in its successor MODAP. Our focus has always been how to reconcile application ideas with data privacy.

Project duration:

September 2009 – February 2013

Link to the project page