LENDÜLET

Society has reached a point of no return, one that leaves us completely reliant on omnipresent ICT-mediated communication. Mobile and sensor-rich portable devices connect millions of humans with petabytes of data and numerous on-line services. However, tearing down the physical-digital barrier in a scalable fashion requires both radically novel algorithmic knowledge and in-depth understanding of humans and societies. We will deliver major theoretical advances in real-time intelligent information management of large datasets including online social networks, mobile devices and humans in physical space by delivering three functions: “alert”, by real-time location-aware knowledge acquisition, analysis and visualization; “response”, through on-demand composition and coordination of large teams; and effective “communication”, through recommendation and personalization.

“Big Data” is an emerging new research area for the methodologies of extreme large scale problems in business intelligence, e-science and Web mining. We concentrate on applications for social network mining, graph clustering, personalized and similarity search, recommendation and spam filtering, as well as security problems ranging from financial risk analysis or insurance fraud to people trafficking or organized crime.

We plan to conduct research ranging from theory to experimentation by building on the unique nature of our research group. We cover the full chain from core research to industrial deployment, including unique access to data ranging from telecommunication logs to large scale Web crawls. As a particular strength in our previous results, we design algorithms that handle the explosive growth in data sizes and impose no artificial size limits for real-world applications. The highlights of our proposed research with both novel areas as well as related fields where we have the strongest existing results are:

* Application of distributed and many-core computing over large scale
   problems of information retrieval and business intelligence;
 * Data mining, machine learning and efficient algorithms applied for
   quality assessment, fraud detection and security;
 * Multimedia information retrieval and recommendation over new
   platforms (mobility, fusion of Internet and television, Web 3.0).
 * Data driven mathematical, algorithmic and data mining modeling
   approaches in human-machine coordination.