קבוצות ונושאים

Department of Software Engineering & Information systems

DATA MINING, TEXT MINING

Renata Avros, Zeev Barzily, Mati Golani, Katerina Korenblat, Dvora Toledano-Kitai, and Zeev Volkovich
Data mining deals with data understanding and data interpretation presented as usual in large datasets. The main research topic in this area is the cluster validation problem. Resting upon modern statistical and computational tools, several new cluster stability criteria have been offered together with applications to text mining problems, like authorship recognition and themes selection.

BIOINFORMATICS

Alex Bolshoy, Katerina Korenblat and Zeev Volkovich
The research in this area is mainly concentrated on two directions. The first concerning the organisms’ classification using properties of DNA sequences. The second deals with feature selection in the
microarray data arrangement. Particularly, the problem of active genes selection in Leukemia diagnoses
is considered.

SOFTWARE ENGINEERING, REQUIREMENTS ENGINEERING, SYSTEM SPECIFICATION AND DESIGN METHODOLOGIES

Zeev Barzily, Mati Golani, Eliezer Kantorovich and Avi Soffer
Software engineering is concerned with the study and application of a systematic, disciplined and quantifiable approach to development; operation and maintenance of software. It involves both computer science and engineering principles and practices towards improved handling of requirements, specification, design, implementation and verification of software systems. It is also incorporates database design and optimization, as well as architecture design ethodologies.
Software Engineering research involves developing applying technologies and practices from computer science, project management, engineering application domains, interface design, digital asset management and other such fields.

GENERALIZED CONVOLUTION THEORY

Renata Avros, Dvora Toledano-Kitai and Zeev Volkovich
Generalized convolution theory is a notable area of modern probability theory. A new analytical approach in this field has been offered by Volkovich. Apparently, one of the most known realizations of this construction is named “Volkovich Convolution”. Recent research studies are focused on the search for new analytical tools that will provide a solution of the convolutions embedding problem.

INFORMATION THEORY AND CODING THEORY

Biniamin Mounits, Zeev Volkovich and Orly Yahalom
The research in this area is concerned with finding upper bounds for the maximal cardinality of codes of given block length with respect to the minimum hamming distance over finite fields. We search for
bounds for the minimum average distance for subsets in metric association schemes. In addition, we wish to develop construction methods for constant weight codes and algebraic geometry codes with improved parameters.

APPLICATION OF KNOWLEDGE ENGINEERING, MODEL-DRIVEN SYSTEMS AND SOFTWARE ENGINEERING
 
Zeev Barzily, Mati Golani, Eliezer Kantorovich and Avi Soffer
Knowledge Engineering is an engineering discipline that involves integrating knowledge into computer systems in order to solve complex problems, normally requiring a high level of human
expertise by building, maintaining and development of knowledge-based systems. Knowledge engineering has a great deal in common with software engineering, and it is related to many computer science domains such as artificial intelligence, database systems, data mining, expert systems, decision support systems and management information systems. Model-driven engineering (MDE) is a software development methodology which focuses on creating models, or abstractions, related to particular domain concepts. It is meant to increase productivity by maximizing compatibility between systems, simplifying the process of system development and maintenance, and promoting communication between individuals and teams working on the system.
 

TRAJECTORY PLANNING FOR MULTI AGENT SYSTEMS

Miri Weiss-Cohen, Gidon Avigad (Department of Mech. Engineering) and Erela Eisenstadt (Department of Mech. Engineering)
Planning free collision trajectories for cooperating multi agent systems, by using different evolutionary approaches. The research involves the optimization of agents’ trajectories, which are subjected to reciprocal constraints, such as: distance between the agents lesser than a predefined distance, keeping the distance between the agents greater than a predefined distance (avoiding collisions); always allow an eye of sight contact between the agents; always occupy one agent at each sub-workspace. The research includes developing new tools of combining B- Splines, NURBS curves and genetic algorithms.

SCULPTURED SURFACES COVERAGE OF 3D SOLIDS USING GA

Miri Weiss-Cohen
NURBS Surfaces for 3D solids, as the solid is arranged on a flat surface. The research aims to optimize the residue of non covered surface, and the gap between the covering 3D solid and the sculptured surface. This optimization involves using Genetic Algorithms for defining the optimal mesh for the surface.

WIRELESS SENSOR NETWORKS

Peter Soreanu and Zeev Volkovich
Wireless sensor networks-related research is a hot topic, due to technological advances that allow for new and complex applications. Performance improvements, mostly energy, efficiency and security related, are of outmost importance in this field. The main research topic concentrates on energy-efficient routing, sensing, data fusion and communication algorithms. Other related topics in which we currently actively research are: efficient clustering, mobile sensors, and security and tracking issues.

RECONFIGURABLE ARCHITECTURES

Esti Stein and Yosi Ben-Asher (Haifa University)
The research mainly involves developing algorithms for existing architectures such as FPGA, and future architecture models such as Reconfigurable Mesh. These architectures have parallel components, and are mostly used to accelerate the execution time of programs or algorithms.

COMPILER OPTIMIZATIONS

Esti Stein and Yosi Ben-Asher (Haifa University)
New hardware architectures (such as VLIW, FPGA, PIPELINE etc.) require new code optimizations for parallelizing the compiled code. This research involves developing algorithms to exploit the new architectures’ parallel components efficiently.