Exploratory Data Analysis with MATLAB

The Third Edition of Exploratory Data Analysis with MATLAB is an intuitive and easy-to-read book that provides practitioners of EDA with the understanding and tools to do EDA. It presents EDA methods from a computational perspective and uses numerous examples and applications to show how the methods are used in practice. The MATLAB code for examples, data sets, and the EDA Toolbox are available for download on the book's website.

Format: Paperback / softback
Length: 590 pages
Publication date: 30 September 2021
Publisher: Taylor & Francis Ltd

The authors have crafted a comprehensive and user-friendly book that serves as an excellent guide for those embarking on the journey of Exploratory Data Analysis (EDA). Accompanied by numerous examples, proposed exercises, comprehensive references, and extensive appendices, this text effectively introduces readers who are unfamiliar with MATLAB to the world of EDA.

Adolfo Alvarez Pinto, a prominent figure in the field of International Statistical Review, praises the second edition of the book, stating that it presents an intuitive and easy-to-read approach, making it an invaluable resource for practitioners of EDA who utilize MATLAB. He further emphasizes that the authors have accomplished an exceptional feat by consolidating a vast array of EDA routines, but their true achievement lies in imparting the knowledge and tools necessary to perform EDA effectively.

David A Huckaby, a respected member of the MAA Reviews, also commends the book, highlighting its significance in the realm of data analysis. He emphasizes that EDA plays a vital role in the data analysis process, particularly as computational power and data sets continue to evolve in size and complexity. This text serves as a valuable toolkit for data scientists, equipping them with the methods necessary to effectively visualize and summarize data, generate hypotheses, and develop models.

Exploratory Data Analysis with MATLAB, Third Edition, takes a computational perspective and employs a wide range of examples and applications to demonstrate the practical application of EDA methods. The authors employ MATLAB code, pseudo-code, and algorithm descriptions to elucidate the concepts, ensuring a clear understanding for readers. Additionally, the MATLAB code for examples, data sets, and the EDA Toolbox is readily available for download on the book's website, further enhancing the accessibility and convenience of the material.

In the third edition, the book introduces several exciting new features and advancements. These include:

Random Projections and Estimating Local Intrinsic Dimensionality: The authors delve into the realm of random projections, a powerful technique for dimensionality reduction and data visualization. They discuss various methods for estimating local intrinsic dimensionality, such as principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE). These methods offer valuable insights into the structure and patterns within data sets.

Deep Learning Autoencoders and Stochastic Neighbor Embedding: The book explores the use of deep learning autoencoders and stochastic neighbor embedding in EDA. These techniques, derived from the field of machine learning, have gained popularity for their ability to learn complex data representations and visualize high-dimensional data. The authors demonstrate how these methods can be applied to various tasks, such as dimensionality reduction, clustering, and anomaly detection.

Minimum Spanning Tree and Additional Cluster Validity Indices: The authors introduce the concept of minimum spanning trees and discuss various cluster validity indices, such as the Davies-Bouldin index and the Silhouette coefficient. These indices provide measures of cluster quality and help assess the effectiveness of clustering algorithms. The book provides practical insights into how these indices can be used to evaluate and improve the performance of clustering algorithms.

Kernel Density Estimation: The book introduces kernel density estimation, a statistical method for estimating the probability density function of a continuous variable. It is widely used in EDA to visualize data distributions and identify outliers or unusual patterns. The authors provide detailed explanations and examples of kernel density estimation using MATLAB, showcasing its versatility and effectiveness in data analysis.

Plots for Visualizing Data Distributions: The book includes a comprehensive chapter on visualizing data distributions, such as beanplots and violin plots. These plots provide a visual representation of the distribution of variables within a data set, allowing data scientists to identify trends, outliers, and clusters. The authors demonstrate how to create and interpret these plots effectively, enhancing the interpretability and understanding of data.

Visualizing Cat: The book includes a chapter dedicated to visualizing categorical data, particularly in the context of machine learning. The authors discuss various techniques for visualizing categorical variables, such as categorical plots, heatmaps, and decision trees. These methods are essential for understanding the relationships between categorical variables and predicting outcomes in machine learning applications.

In conclusion, Exploratory Data Analysis with MATLAB, Third Edition, is a highly recommended book for practitioners of EDA who seek to leverage the power of MATLAB in their data analysis endeavors. The authors have crafted a comprehensive and user-friendly guide that provides a solid foundation for understanding and applying EDA methods. With its extensive examples, practical applications, and comprehensive appendices, this text serves as an invaluable resource for data scientists, researchers, and students alike. Whether you are a novice or an experienced data analyst, this book will undoubtedly enhance your skills and knowledge in EDA.

Dimension: 234 x 156 (mm)
ISBN-13: 9781032179056
Edition number: 3 New edition