IBUSCA: A Grid-based Bottom-up Subspace Clustering Algorithm

The paper presents the bottom-up subspace clustering approach and discusses some drawbacks of clustering methods in broad analysis of complex, high-dimensional data. The aim of this paper is to propose some improvements of existing bottom-up subspace clustering methods. A novel grid-based bottom-up subspace clustering algorithm is presented which is able to handle both numerical and nominal attributes and requires only one single parameter. Clusters are represented as hyper-rectangles in sub-spaces of attributes and can be easily interpreted by a human as decision rules. The results of experiments conducted on artificial and real data sets are included