Data Mining and Exploration

This book will introduce both conceptual and procedural aspects of cutting-edge data science methods, such as dynamic data visualization, artificial neural networks, ensemble methods, and text mining. There are at least two unique elements that can set the book apart from its rivals. Most students in social sciences, engineering, and business took at least one class in introductory statistics before learning data science. However, usually these courses do not discuss the similarities and differences between these two schools of thought, and as a result learners are disoriented by this seemingly drastic paradigm shift. In reaction, some traditionalists reject data science altogether while some beginning data analysts employ data mining tools as a black box", without a comprehensive view of the foundational differences between traditional and modern methods (e.g. dichotomous thinking vs. pattern recognition, confirmation vs. exploration, single method vs. triangulation, single sample vs. cross-validation...etc.). To remediate this problem, this book will provide the readers with the details of the similarities and differences between classical methods and data science, as well as the path for the transition (e.g. from p value to LogWorth, from resamplingto ensemble methods, from content analysis to text mining...etc.)"--