Data Science. Science of Data from Scratch

The book allows you to study the science of data (Data Science) and apply the acquired knowledge in practice. It contains a brief course on the Python language, elements of linear algebra, statistics, probability theory, and data processing methods. The basics of machine learning are introduced. Algorithms of k-nearest neighbors, naive Bayes classification, linear and logistic regression, as well as models based on decision trees, neural networks, and clustering are described. Techniques for processing natural language, methods for analyzing social networks, the basics of databases, SQL, and MapReduce are discussed. In the second edition, the examples have been rewritten in Python 3.6, toy datasets have been replaced with 'real' ones, materials on deep learning and data ethics, statistics and natural language processing, recurrent neural networks, word vector embeddings, and matrix decomposition have been added.