Automatic classification of data-warehouse-data for information lifecycle management using machine learning techniques

The aim of Information Lifecycle Management (ILM) is to govern data throughout its lifecycle as efficiently as possible and effectively from technical points of view. A core aspect is the question, where the data should be stored, since different costs and access times are entailed. For this purpose data have to be classified, which presently is either done manually in an elaborate way, or with recourse to only a few data attributes, in particular access frequency. In the context of Data-Warehouse-Systems this article introduces an automated and therefore speedy and cost-effective data classification for ILM. Machine learning techniques, in particular an artificial neural network (multilayer perceptron), a support vector machine and a decision tree approach are compared on an SAP-based real-world data set from the automotive industry. This data classification considers a large number of data attributes and thus attains similar results akin to human experts. In this comparison of machine learning techniques, besides the accuracy of classification, also the types of misclassification that appear, are included, since this is important in ILM.