Typically discretisation procedures are implemented as a part of initial pre-processing of data, before knowledge mining is employed. It means that conclusions and observations are based on reduced data, as usually by discretisation some information is discarded. The paper presents a different approach, with taking advantage of discretisation executed after data mining. In the described study firstly decision rules were induced from real-valued features. Secondly, data sets were discretised. Using categories found for attributes, in the third step conditions included in inferred rules were translated into discrete domain. The properties and performance of rule classifiers were tested in the domain of stylometric analysis of texts, where writing styles were defined through quantitative attributes of continuous nature. The performed experiments show that the proposed processing leads to sets of rules with significantly reduced sizes while maintaining quality of predictions, and allows to test many data discretisation methods at the acceptable computational costs.

The performance of a classification system of any type can suffer from irrelevant or redundant data, contained in characteristic features that describe objects of the universe. To estimate relevance of attributes and select their subset for a constructed classifier typically either a filter, wrapper, or an embedded approach, is implemented. The paper presents a combined wrapper framework, where in a pre-processing step, a ranking of variables is established by a simple wrapper model employing sequential backward search procedure. Next, another predictor exploits this resulting ordering of features in their reduction. The proposed methodology is illustrated firstly for a binary classification task of authorship attribution from stylometric domain, and then for additional verification for a waveform dataset from UCI machine learning repository.

Knowledge discovered from data can be represented in a form of decision rules, consisting of required conditions and decisions to which they lead. The quality of rules is usually considered in terms of some quantitative measures such as confidence, support or length. Depending on all these parameters the constructed classifiers can greatly vary in the predictive accuracy and the size of their structure. Both these elements depend strongly on the choice of characteristic features, which can be found by some independent feature selection procedure, but also by applying a wrapper model. In the wrapper model the classifier and its parameters are used to evaluate the importance of attributes. In the paper there are proposed measures of attribute relevance based on rule lengths. The usefulness of the described methodology is shown for rule-based classifiers, obtained through Dominance-Based Rough Set Approach, and a connectionist solution implemented with Artificial Neural Networks, both employed in the task of authorship attribution.