Using clementine data mining system in the process of analysis of cotton fiber properties

Abstract

In the last years, great developments in technology have taken place in certain branches of testing procedure. Measures like micronaire, length, uniformity, strength, elongation, colour and trash contents are determined easily using the High Volume Instruments Systems (HVI), providing rapid and reliable results. However, cotton chemical properties are obtained using laboratory methods that are more time consuming. This knowledge is very important because chemical properties can affect wet processes. Several studies were made relating, physical properties and yarn characteristics, physical properties between each others or between instruments used in quality control. All of these studies used statistical tools to achieve those relations. This paper shows a preliminary study made in a research project named Cotton Properties: Inference through Data Mining Techniques founded by Science and Technology Foundation . In this project are used several Data Mining techniques, available in the Clementine Data mining System. Data Mining constitutes one of the steps of the Knowledge Discovery (KDD) process, a process that aims the discovery of associations within data sets. Clementine includes advanced modelling techniques, like machine learning technologies, wich extract complex relationships and decision-making rules from the data. These help to automate applications such as prediction, estimation or classification, and can be used to provide “expert” decision support [4]. This paper describes the use of Clementine in the analysis of a database storing physical and chemical properties of cotton fibres. The results achieved point out that Data Mining techniques can effectively be used in the establishment of the relationships that characterise cotton fibre properties.