Overview

Data science is an interdisciplinary field that uses methods, algorithms, processes, and systems to extract knowledge and conclusions from all types of data. Through machine learning, it combines elements of statistics, computer science, mathematics, and analysis techniques to solve problems, make predictions, and generate value from data. It relies on large volumes of data (big data) to discover patterns, trends, and relationships that can be used for decision making. Machine learning uses two types of techniques: supervised learning, which trains a model with known input and output data so that it can predict future outcomes, and unsupervised learning, which finds hidden patterns or intrinsic structures in the input data. This book develops most of the unsupervised learning techniques. It begins by delving into dimension reduction techniques, such as principal component analysis and factor analysis, which aim to eliminate the harmful effects of correlation in analyses with quantitative variables. It then discusses simple and multiple correspondence analysis to relate categorical variables to each other. The following chapters focus on classification and segmentation using cluster analysis, taking into account hierarchical and non-hierarchical techniques. Special emphasis is placed on the usefulness of dimension reduction prior to segmentation. Next, segmentation is addressed through multidimensional scaling, using metric and non-metric scaling and highlighting its usefulness in the analysis of preferences and in geographic positioning, which is very useful in logistics. Finally, advanced topics are developed, such as cluster analysis through appropriate neural networks. First, the organizational map networks (SOM) such as the Kohonen network are considered, and then the Autoencoder neural networks are used, which are common tools in the development of Deep Learning. Another advanced topic that is also developed in this book is pattern recognition through neural networks such as the Hopfield network and convolutional neural networks (CNN) for image processing. Anomaly detection is also discussed. For all topics, methodological concepts are presented and illustrated with practical examples and exercises fully solved in Python code.

Full Product Details

9798340962041

Table of Contents

Reviews

Author Information

Tab Content 6

Customer Reviews

Recent Reviews

Countries Available