Unsupervised Learning

This article presents a review of traditional and current methods of classification in the framework of unsupervised learning, in particular cluster analysis and self-organizing neural networks. Both are vector quantization methods aiming at minimizing the distance between an input vec- tor and its representation. The learning is unsupervised as no predefined cluster structure of the input data is as- sumed. The review of cluster analysis methods covers hard clustering, hierarchical and nonhierarchical, whose aim is to assign exact (with membership degree equal to 1) units (objects) to clusters; fuzzy clustering, where the member- ship degree of a unit to a cluster is allowed to stay in the interval [0; 1]; mixture clustering, a model-based clus- tering consisting in fitting a mixture model to data and identifying each cluster with one of its components. All these methods are reviewed in all the variants related to the presence of complex or big data structures or to the presence of outliers. The self-organizing maps are also presented as artifi- cial neural network, the cells (neurons) of which become specifically tuned to various input data patterns or classes of patterns through an unsupervised learning process. The resulting vector