The technology of knowledge-based system construction by knowledge extraction from examples has been successfully employed in a number of practical studies. Wide application of inductive algorithms can be seen in forecasting, especially as applied to medicine (assistance in choosing treatment method or therapy) as well as weather [2], [3] and financial market forecasting and pattern recognition [4]. This paper presents the results of diagnosing some condition attributes at spontaneous intracerebral extravasations (haematomas). Section 2 provides a description and motivation of choosing the subject area. Since all the experiments are performed using a group of inductive methods (CART, ID3 and CART2), they are also with in outline. The CART2 algorithm is described in more detail as it is less known. This algorithm, however, produced the best results during the experiment. In further sections, working with data is described and a comparative table is given representing the results of all algorithms execution. Finally, an analysis of the results obtained is presented.
[1]
Stan Matwin,et al.
Machine Learning for the Detection of Oil Spills in Satellite Radar Images
,
1998,
Machine Learning.
[2]
Sholom M. Weiss,et al.
An Empirical Comparison of Pattern Recognition, Neural Nets, and Machine Learning Classification Methods
,
1989,
IJCAI.
[3]
K. Vemmos,et al.
DIAGNOSIS OF STROKE USING INDUCTIVE MACHINE LEARNING
,
1999
.
[4]
C. A. Lindley,et al.
Toward building an expert system for weather forecasting operations
,
1994
.
[5]
P. Guignard,et al.
Performance Comparison of Different Learning Methods for Weather Forecasting Operations
,
1995
.