LUMII

LUMII > Computer Science > Main research directions

Main research directions

Complex systems modelling languages and tools development

The research in this area continues from 1980s. The most important achievement is one of the most up-to-date in the world system modelling tool GRADE (created in cooperation with German company Infologistik GmbH 1990 -2000). GRADE is still unmatched system for managing very large graph-like diagrams. Research goes on with the most contemporary software design approach - Model Driven Architecture (MDA), Transformation Based Graphical Tool Building Platform. Special languages for model transformation and transformation based tools aimed at complex system modelling and design are being developed.

Graph theory and visual information processing

Image processing, analyze and machine vision systems; graph theory, algorithms and graph drawings are important research areas for the Institute. Research is being conducted at several levels: theoretical (topological graph theory) and practical (graphs in bioinformatics). The development of rain sensor (automatically controlled cleaning equipment for a plate surface) and fake coins detector are the most important successes in the field of visual information processing. Researches in graph drawing helped to create Graph-Drawing tools for GRADE and Exigen Business Modeler. Successful cooperation has lasted for years with Tom Sawyer Software Corporation (USA). The Institute creates graph layout algorithms for company.

Semantic web technologies

Next generation’s information systems, based on semantic technologies, may soon replace the traditional information systems, based on RDBMS. The main goal of semantic methods is to annotate information found in data bases and text storages with a help of formal ontological tools so it would be „understood” not only by humans but also by automatic computer programs. The achieving of this goal would provide possibilities to sort information in fundamentally new ways, basing on semantics. One of the most important steps to this goal is to set up national semantics infrastructure in time. The most important elements of infrastructure are described in vision called “Semantiskā Latvija” (“Semantic Latvia”). It schedules creation of ontological portal and centralized RDF data storage and preparation of ontological and linguistic resources specific for Latvia. The most important project in research of this field is „SemTi Kamols” .

Computational linguistics

Computational linguistics (CL) central problem is analysis, representation and synthesis of a meaning expressed in a form of natural language. A number of complicated problems in lower levels of language analysis (namely in morphology, syntax and semantics) have to be solved in advance. There are several directions of developing Latvian language resources and language managing tools at Institute: development of morphological and syntactical analyzers, creation of machine-readable dictionaries and lexical onthologies, text corpora as accumulation of empirical knowledge, modeling of discourse representations, rule-based and statistical machine translation, speech synthesis and recognition.

Bioinformatics

There is active ongoing research in several theoretical areas of bioinformatics, most importantly: development of methods for analysis/comparison of protein structures as well as methods for studies on structure evolution; modelling of gene regulatory networks and development of methods for analysis of network behaviour, stability etc. as well as methods for modelling gene network evolution.
On a more practical level bioinformatics group specializes in development of advanced software and database systems for storage and analysis of biomolecular data obtained from high-throughput technologies. One of the largest projects in this field is SIMBioMS: a web-based software system developed for collection, storage, management and retrieval of data and information about patients and biomedical samples as well as experiment results obtained using various platforms. SIMBioMS initially was developed as a data exchange system for use in EU FP6 project MolPAGE (main emphasis on studies of type 2 diabetes and other metabolic diseases) and has been later adapted for several other applications, including EU FP7 project ENGAGE.

Real time systems

The Institute has 15 years of experience in the area of telecommunications software contracts (wireles telephone acces systems; radio broadcasting studio control systems; integration of video surveillance, allarms and telecomunication devices). The researches in this area will continue. Institute has planned to join activities of European Union FP7 programme’s embedded systems platform.

Mathematical foundations of computer science

In this area the research focuses on quantum algorithm theory: the methods of evaluating algorithms complexity will be developed and search for efficient quantum algorithms will continue. More efforts will be paid to develop inductive synthesis theory (to induce regularity from stray instances).