Algorithms for Data Science

Más información sobre el libro

This textbook on practical data analytics focuses on the essential principles, algorithms, and data. Algorithms are central, with clear explanations of their mathematical and statistical foundations. However, effective data analytics goes beyond theory; it requires programming fluency and hands-on experience with diverse data. Readers engage with Python and R, developing the ability to adapt algorithms for innovative analyses. The book is divided into three parts: (a) Data Reduction introduces concepts like data maps and information extraction, followed by associative statistics and scalable algorithms, including practical aspects of distributed computing with Hadoop and MapReduce. (b) Extracting Information from Data covers linear regression and data visualization, with a dedicated chapter on Healthcare Analytics that utilizes the CDC's Behavioral Risk Factor Surveillance System, appealing to practitioners working with large datasets. (c) Predictive Analytics details foundational algorithms like k-nearest neighbors and naive Bayes, includes a chapter on forecasting, and explores streaming data with tutorials using Twitter API and NASDAQ data. Designed for upper-division undergraduate and graduate students in mathematics, statistics, and computer science, the book requires minimal prerequisites. Each chapter is accessible, with exercises of varying difficulty, making it suitable for self-study and a valuable resource