Study in Zamość
Pearl of Renaissance architecture
One of the 7 Wonders of Poland
University of Management and Administration in Zamosc
Akademicka Street 4, 22-400 Zamosc, Poland
tel./fax +48 84 677 67 09, 677 67 10
Polski Ukrainian

Forecasts for SME sector

Research focused on the economic situation and forecast of SME  sector was carried out  by the scientific team from The Zamosc University of Management and Administration,  supported by experts from The Wroclaw University of Economics and the University of Information Technology and Management in Rzeszow

Econometric methods

Surveys include two main issues:

  • Creating a standard (benchmark) of a small administrative region for Lubelskie and Podkarpackie Voivodeship
  • Development of forecasting models to allow a prediction of economic and financial viability of enterprises in the future

Artificial intelligence methods

The main area of research includes:

  • Using Bayesian Networks for clarification of complex objects behaviour
  • Construction models for forecasting this behaviour with using existing tools and comparing the obtained results with application of different methods of artificial intelligence (bayesian networks, neural networks, decision-making tree).

The subjects of the study are small businesses and their economic  and financial condition

Unemployment prevention system in underdeveloped areas


Andrzej Burda, Zdzisław S. Hippe:

Uncertain Data Modeling: The Case of Small and Medium Enterprises

Abstract: A new procedure for combined validation of learning models – developed for specifically uncertain data – is briefly described; it relies on a combination of resubstitution with the modified learn-and-test paradigm, called by us the queue validation. In the initial experiment the elaborated procedure was checked on doubtful (presumably distorted by creative accounting) data, related to small and medium enterprises of the Podkarpackie-region in Poland. Validated in the research learning models were completed in the form of decision trees and sets of production rules. Correctness of both types of models (trees and rules) was estimated basing on the error rate of classification. It was found that false-positive classification errors were significantly larger than false-negative ones; the difference discovered by validation procedure can be probably used as a hint of fraud in the evaluated data.

Keywords: validation, small and medium enterprises, creative accounting

This paper appears in: Human System Interactions (HSI), 2010 3rd Conference on
Issue Date: 13-15 May 2010
On page(s): 76 – 80
Location: Rzeszow
Print ISBN: 978-1-4244-7560-5
INSPEC Accession Number: 11445829
Digital Object Identifier: 10.1109/HSI.2010.5514586
Date of Current Version: 23 lipiec 2010
Link to article:

Andrzej Burda, Zdzisław S. Hippe:

Methodology of Predicting the Future Condition of Small and Medium Enterprises

Abstract: Our paper presents the concept behind a new methodology of predicting the future condition (i.e. over-all state) of small and medium enterprises, relied on results of a specific data mining procedure. The developed procedure is based on the controlled improvement of the source data set (describing the investigated objects) with the use of an ensemble of artificial neural networks, EANN. Then, in the next step of the procedure, various machine learning algorithms are applied to generate models that figure out inherent properties of the investigated objects. The reliability of the prediction, confirmed in a very exhaustive tests conducted on data from the span of 8 years, was close to 80% of correctness.

This paper appears in:
in Polish: Regional Barometer, No 1 (19) 2010,
Zamość University of Management and Administration, Zamość 2010
Link to article:

in English: Tadeusiewicz, R.; Ligęza, A.; Mitkowski, W.; Szymkat, M. (Eds.) 
7th Conference Computer Methods and Systems,
Oprogramowanie Naukowo Techniczne, Kraków:61-64

B. Kuczmowska, A. Burda, Z. Hippe:

Prediction of Economic Situation of Small and Medium Enterprises Using Bayesian Network

Summary: The structure of a belief network model for predicting the economic situation of small and medium enterprises in the Lubelskie and Podkarpackie provinces is discussed. The direction of further scientific research in order to improve performance of a belief network model is additionally presented.

M. Kurzynski, E. Puchala, M. Wozniak, A. Zolnierek,  (Eds.)
Computer Recognition Systems 2,
Springer, Berlin/Heidelberg, 2007, pp. 802-807
Link to article:

A. Burda, B. Kuczmowska, Z. Hippe:

Ensembles of Artificial Neural Networks for Predicting Economic Situation of Small and Medium Enterprises,

Summary: Results of predicting an economic and financial situation of small and medium size enterprises from Lubelskie and Podkarpackie regions are presented. As opposed to other research in this field, the micro-macro (mezzo-) modeling concept was used. The results of prediction using neural perceptions and two types of ensembles of artificial neural networks were compared. Some conclusions and directions of further research aiming at improving quality and extending the functionalism of enterprises being investigated are dealt with.

M. Kurzynski, E. Puchala, M. Wozniak, A. Zolnierek,  (Eds.)
Computer Recognition Systems 2,
Springer, Berlin/Heidelberg, 2007, pp. 808-815
Link to article:


gallery-eng-12 gallery-eng-16 gallery-eng-18 gallery-eng-20 gallery-eng-26 gallery-eng-4 gallery-eng-7 gallery-eng-8