esearch areas: S. Velickov Machine learning and statistical learning theory Application of Machine Learning (ML) in Hydroinformatics through the development of algorithms that can automatically learn from quantitative and qualitative data-that is, build models, extract patterns, and adapt to changing situations-and making software tools for human data analysts to more easily explore and better understand data based on statistical learning theory. My main research interests include: induction algorithms using decision trees and progressive rule generation; sub-symbolic learning methods such as Artificial Neural Networks and other connectionists models. Support Vector Machines (SVM) based on Vapnik's statistical learning theory. concept hierarchies and their automatic generation. automated knowledge acquisition ("learning to learn"). Data Mining and Data-driven modelling The large potentials in the existing data banks related to water resources and environmental engineering need to be explored in order to transform these data/observables into valuable information and knowledge. The key to these potentials can be found in data mining, which as a new emergent field in hydroinformatics provides the technologies and tools to describe, understand and predict water-related and environmental natural processes. These techniques are divided into two general categories: (a) prediction and (b) knowledge discovery. The focus of my research in this field is put on predictive data mining. The main goal of the predictive data mining is through processing of data/observables (records from the past) to describe the underlying dynamics of the complex systems and to discover hidden knowledge (such as: clusters, patterns, associations, rules, correlation properties, trends) from those data. Furthermore, these predictive data mining techniques have the possibilities to reconstruct the dynamics of the systems from the observables and predict the future behavior of the system. My main research interests include: exploratory data analysis. wavelet analysis of time-series. clustering and classification. association and classification rule mining. local non-linear modelling of time-series based on chaos theory. knowledge mapping and visualization. Internet technologies and computing Internet and intranet environments provide exciting opportunities for remote modelling, collaborative engineering and efficient knowledge management. Here at hydroinformatics section we are exploring the emergent internet technologies that may be used to build net-based applications in hydroinformatics, such as CGI, ISAPI/NSAPI, ActiveX, Java, ASP, XML, ColdFusion, Flash and other technologies. We study their philosophy, structure, advantages and disadvantages, and their potential use for remote modelling, distributed computing (and programming), distance learning, and collaborative design and decision making processes over the Internet. Implementing the above-mentioned technologies, several prototypes were built as remote modelling hydroinformatics systems, mainly for an educational and distance learning purposes. Knowledge Management Shortly, Knowledge Management can be defined as "theory and strategy based upon the synergy between human creativity and capabilities afforded by new information technologies". As a member of the Knowledge Management team within the Delft Cluster project, my main research interests in this area include: knowledge sharing and distribution. document and content management. text mining. collaborative engineering and communities of practice. knowledge maps: automated generation and navigation. knowledge portals. Multiagent systems and Distributed Artificial Intelligence Distributed Artificial Intelligence (DAI) today is established and promising research and application filed which brings together and draws on results, concepts and ideas from many disciplines, including Artificial Intelligence (AI), computer science, sociology, organization and management science, and psychology. Its broad scope and multi-disciplinary nature make it difficult to precisely define DAI in a few words. One possible definition could be that DAI is study, construction and applications of multiagent systems. Multiagent systems are systems in which several interacting, intelligent agents perform some set of tasks. As hydroinformtics has very strong socio-technical dimension there are many common points with DAI and challenging research issues that need to be explored. My main research interests in this area include: Instantiation of simulation models using multiagent systems. Distributed computing using multiagent systems. Distributed decision-making using multiagent systems.