Loading...
Projects / Programmes source: ARRS

Computerised methods in the study of relationships between systems’ parameters

Research activity

Code Science Field Subfield
2.07.07  Engineering sciences and technologies  Computer science and informatics  Intelligent systems - software 

Code Science Field
P175  Natural sciences and mathematics  Informatics, systems theory 
P176  Natural sciences and mathematics  Artificial intelligence 
Keywords
QSAR, bio and photochemical degradation, pollutants, collaborative discrete simulation
Evaluation (rules)
source: COBISS
Researchers (8)
no. Code Name and surname Research area Role Period No. of publicationsNo. of publications
1.  05685  Marko Ahčan  Engineering sciences and technologies  Researcher  1997 - 1999  19 
2.  04855  MSc Janez Benkovič  Engineering sciences and technologies  Researcher  1997 - 1999  59 
3.  12090  PhD Danica Dolničar  Engineering sciences and technologies  Researcher  1998 - 1999  170 
4.  04810  PhD Saša Aleksej Glažar  Social sciences  Researcher  1997 - 1999  1,228 
5.  03183  MSc Dragotin Kardoš  Natural sciences and mathematics  Researcher  1997 - 1999  135 
6.  12091  PhD Aleš Musar  Engineering sciences and technologies  Researcher  1997 - 1999  107 
7.  08957  PhD Margareta Vrtačnik  Natural sciences and mathematics  Principal Researcher  1999  866 
8.  16012  PhD Katarina Senta Wissiak Grm  Social sciences  Researcher  1997 - 1999  266 
Organisations (1)
no. Code Research organisation City Registration number No. of publicationsNo. of publications
1.  1555  University of Ljubljana, Faculty of Natural Sciences and Engeneering  Ljubljana  1627074  20,201 
Abstract
The proposed project is a contribution to the methodology of knowledge engineering in identifying a problem, selecting and evaluating parameters and interpretation of the results of methods used in computer supported studies of the relations among the parameters in various processes. The prediction and accuracy of the models for defining the relationships among parameters depend to a great extent on: (1) selection of data teaching and testing sets (examples of compounds and/or processes) for studying relationships, (2) selection and hierarchy of parameters for the description of data set entities, (3) selection of the computer method for identifying relations with regard to the problem definition, (4) iterpretability of results, (5) generalisation of the results to suit relevant types of problems. The assessment methodology is tested on two types of processes: (1) studying interactions of pollutants in the environment and (2) prediction of mass flows in chemical and related processes. For studying the interactions of pollutants with the environment, classification models are employed for the prediction of the relations among electronic and topological parameters and for the properties of substances, algorithms ID3, CART, C4.5 and SIPINA are used. In addition, experimental studies of mechanisms of bio and photocatalytic biodegradation in correlation with the structure of model compounds have been made. For the development of a model for predicting mass flows, a model with collaborative discrete event model-network has been designed, in which the central model (e.g. environmental legislation) has the additional role of synchronising the activities of other models in the network.
Views history
Favourite