L3S Research Center -  RSS Feeds

Sie sind hier


Is there a new financial crisis on the horizon? This is a very big and challenging question that many financial institutions, governmental bodies and regulators are trying to answer. The emerging field of systemic risk analysis aims to answer this question by analyzing and monitoring the complex co-dependency networks of market players and their exposure to risk. The goal of the QualiMaster project is to develop a Big Data infrastructure for real-time analysis of large volume, high velocity data streams from the financial markets as well as from the social web in support of effective systemic risk analysis.


The failure to detect the recent global financial crisis early enough has shown that existing analytic models and technologies are not yet capable of accurately detecting such crises or estimating their impact on the global market. The emerging field of systemic risk analysis aims to fill this gap by detecting and managing the exposure to risk of financial market players such as banks, companies or traders and analyzing their complex co-dependency network. However, conducting such a complex analysis in real-time remains a challenge due to the lack of enabling technologies that support the real-time and simultaneous processing of large numbers of data streams at high level of velocity (i.e. high arrival rate of new data).


Challenges & Highlights

The QualiMaster project aims to develop an infrastructure to enable the real-time analysis of high volume and high velocity data streams in support of financial systemic risk analysis. This is realized by building on state-of-the-art Big Data technologies such as Apache Hadoop and Apache Storm. One of the challenges that QualiMaster has to deal with is the dynamicity and volatility in the financial markets, which can lead to significant variances in the data load and quality. QualiMaster handles this through its quality-aware configuration and adaptation model, which enables autonomous quality-driven adaptation of its data processing elements at run-time. One of the distinguishing features of QualiMaster is the integration of re-configurable hardware, such as FPGAs, in its computing model. By translating some of the heavy computations to FPGAs, QualiMaster can accelerate the realtime processing and analysis of data streams and reach high levels of scalability. In addition to the financial data streams, QualiMaster provides scalable and quality-aware algorithms for processing and analyzing web and social media streams, such as event detection, opinion mining and social network analysis. The goal is to use these media as societal sensors for providing additional signals that can complement the financial systemic risk analysis.


Potential Applications & Future Issues

The main application of the QualiMaster project is risk analysis in financial markets. The models, algorithms, tools and the QualiMaster infrastructure developed in the project will be validated and tested using data from the financial domain as well as relevant web and social web content. However, it is also expected that a significant part of the results will be applicable for real-time analysis of data streams in other domains.

Prof. Dr. techn. Wolfgang Neidl
Prof. Dr. tech. Wolfgang Nejdl

Project type:
Specific Targeted Research Project EU/IST FP7
Project duration:
Jan 2014 – Dec 2016
Project research areas:

 Web Information Management

Project manager:
Dr. Mohammad Alrifai