Computational Analytics is the field concerned with computer-based analysis of large amounts of data. Dealing with data of this scale demands (the use of) sophisticated tools. Choosing the right tool for the job requires deep knowledge about the underlying algorithms and data structures, together with comprehensive analyzes to shed a light onto their capabilities and limitations. To further extend these capabilities, new approaches need to be developed, utilizing methods from Algorithm Theory, Machine Learning and Statistics. Additionally our cooperation with researchers and specialists from relevant fields allows for highly beneficial integration of domain-specific knowledge.
Our research focuses on the design, development, theoretical analysis and practical evaluation of new algorithms and data structures for combinatorial optimization problems on structured data that can be modeled as graphs or networks. The methods we use are often based on combinatorial graph algorithms and efficient data structures, that utilize methods from data mining and machine learning on a wide range of applications, which include but is not limited to Decomposition methods, Randomization, Approximation methods, Parameterized Algorithms, structural investigations (Graph Theory), Integer Programming, and Polyhedral Combinatorics. And recently, we are widening our focus to include Parallel Graph algorithms and Data Stream algorithms due to their increasing relevance and applicability.