Data Warehousing and Business Intelligence Group

 
 

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 lab. 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 listed next.

Research areas

• Big data
• Business Intelligence

Selected achievements

The Laboratory hosts the ERC Starting Grant winner Dániel Marx and the winner of the Momentum Grant of the Hungarian Academy of Sciences, András Benczúr. The R&D results of the laboratory focus on data mining and search solutions for community and link analysis, custom solutions for extreme large systems (large Intranets, high traffic portals) as well as for languages with particularly complex syntax in collaboration with computational linguistic groups. For the quality of our research results we were awarded a Yahoo! Faculty Research Grant in the academic year 2006/2007. In 2007 the Group achieved First Prize on the prestigious KDD Cup, a competition involving the best data mining groups around the world. Several of our former PhD students work now at the research centers of the leading internet search companies
Know-how and industrial solutions
Major software products include a customer relation management software capable of visualizing the connection between entities (persons, objects, contracts) as well as in a search engine with integrated linguistic tools for the Hungarian language that serves the Intranet of national branches of multinational companies (T-Mobile, Vodafone, AEGON). We led several customer relation management and risk assessment projects for AEGON Hungary.

Products

• Hungarian Telecom: Intranet and portal search solutions, fully operational since 2004.
• Web server and IT log analytics system, T-Online, AEGON Hungary.
• Hungarian Telecom: Call community and customer analysis tool.
• AEGON Hungary: Desktop search engine, data warehousing, customer network analysis, fraud detection, car insurance campaign toolkit, 2006-.

More information:
Image removed.

Manager

Ph.D.
research fellow