DATA SCIENCE THROUGH PYTHON. UNSUPERVISED LEARNING: CORRESPONDENCES, CLUSTER ANALYSIS, MULTIDIMENSIONAL SCALING AND NEURAL NETWORKS

Data Science is the foundation of Artificial Intelligence and the future of all complex decision-making processes, combining mathematical algorithms and machine learning techniques. Statistical techniques greatly support data science algorithms. Throughout this book, many unsupervised learning techniques are developed from a methodological and practical perspective, with applications using Python software. Classification and segmentation techniques such as Cluster Analysis, Multidimensional Scaling, and Correspondence Analysis are explored in depth. The use of neural networks for classification is specifically developed, addressing Kohonen networks, SOM networks (Self-Organizing Maps), Convolutional Neural Networks (CNNs), Hopfield networks, anomaly detection, autoencoders, and pattern recognition. All techniques are approached from a dual theoretical and practical perspective.