Computerised Decision Making Support for Textile Engineering and Clothing Design

In the present paper we present the application of machine learning in building decision making models used for consuiting about the clothing style, cost range, colour etc. of young people. Such models can be used in designing, planing the production, marketing and fashion forecasting of clothing, or in the textile engineering. The machine learning method used is an extension of a decision tree concept to a multidimensional - vector - decision tree constructed with the help of evolutionary programming. In contrary to the common decision tree the vector decision tree can make more than just one suggestion per input sample. It has the functionality of many separate decision trees acting on a same set of training data and an swering different questions. Vector decision tree is therefore simple in its form, is easy to use and analyse and can express relationships between decisions not visible before (form of data mining). We implemented the above concept in the vector decision trees induction tool called DecRain. Generated vector decision trees showed improvements in comparison to classical decision trees. The accuracy and decision power of decision trees was high, and some intere sting new relation ships has been minded from the database constructed from interviews performed in several secondary schools in the Maribor region