Data-Driven Modeling & Scientific Computation: Methods for Complex Systems & Big Data

I have been reading this book steadily recently and it has repaid my serious study. It is written in a very clever way, focusing on giving the big picture, while not neglecting considerations related to direct application. The concentration on analysis of illustrative examples really aids self-study. The author displays a brilliance in this book in presenting Matlab and interweaving it with the study of important topics in numerical mathematics. The emphasis is on the mathematics, understanding, and applying numerical methods, rather than strictly in technical details of Matlab. I recommend this book as a learning aid for Matlab, if one is interested in certain areas of numerical analysis that the book addresses. The first part of the book is pretty elementary, but the presentation is interesting and engaging, and aspires, to a certain extent, to serve as a nice survey introductory treatment at the high school and freshman college level of applied mathematics. There is a kind of juvenile cuteness about the first several chapters, but the level does start to pick up when differential equations are discussed. Applied mathematics is establishing itself, more and more, as a competitor for traditional pure mathematics, thanks to the power of modern computation. The problems the applied mathematicians work on are growing in significance and originality, due to the impact of improvements in computers and computational techniques. This is the "experimental" or "engineering" side, perhaps, of mathematics, and it can often be as exciting, clever and original as the "theoretical" or pure side of mathematics. My own feeling is that computational science (as we might call this area of mathematics now) hearkens back to the ancient Pythagoreans, and an interest in understanding and grasping the harmony of the cosmos.In the latter portion of the book, much more advanced perspectives are discussed, and I think that the need for real dedicated and persistent effort on the part of the reader (more at the PhD level or a fairly unusually superior undergraduate level of applied mathematics) is needed. At least, I found this to be the case for myself. My background in physics and mathematics was helpful, but I still had to work pretty hard to understand satisfactorily this latter material. To my mind, the most advanced material in the book takes much effort to learn, for a typical American undergraduate student, and a fair background in PDEs, fluid physics, electrodynamics, and quantum mechanics at least at the senior level in physics seems almost a prerequisite for grasping the full extent of this book. Note that the intent of the book is not to transmit rigor and the pure math (albeit it comes with the effort you put in) but to grasp the engineering and applied math aspects, oriented toward physical intuition (which can often be extremely non-intuitive relative to more modern areas of physics). I don't think many undergraduate applied mathematics students will find the material in the latter part of the book that digestible: It seems mostly for grad students and PhD students in applied math intent on going into research. Nonetheless, it is a pretty fascinating journey, if not an easy one.The last few chapters detail possible "real-world" student projects. I thought all of the projects sounded interesting, but I have not done the work needed to explore the problems posed, so this is very subjective on my part.The book is panoramic and explores an exciting vista of the future for applied math applications in data science, engineering and applied math. It is a stunning achievement, and opens an exciting landscape for exploration. It has certainly been one of the best books I have ever read and reminds me a little of Feynman's wonderful three volume intro to physics in its breadth and the brilliance it displays.

Excellent, detailed presentation of Singular Value Decomposition and other essentials of computational data science. Strong emphasis on a groundwork for image processing.Many interesting historical references. Large easy font with great figures and illustrations.Includes a quick/indirect Linear Algebra review; but you really need to have that beforehand (I recommend Gilbert Strang's intro). Also an intro to Differential Equations would be a prerequisite for some later chapters.Some of the math is above my level; and I'll be revisiting those chapters. I learned of Dr. Kutz through his online course, which is great; and this led me to his book.

If you are trying to make use of advanced mathematical tools to analyze your data but you are particularly from engineering background then this is the book. The author did a terrific job in explaining the methods like POD, DMD. There is a short chapter on Finite element analysis too. If you have background on finite volume or other CFD methods but would like to know the core of the Finite Element method this will server that purpose too. Over all I am very pleased to have such a book and I heartily congratulate the author for being successful in his goal of making the mathematical methods accessible to general audience.

I have to admit that I owe Nathan alot. I have listened to his online Data Analysis course, and I was amazed as to how he is able to make dry mathematical concepts be accessible, as well as intuitive.I purchased the book as a companion to the course, and I think that Nathan, should write other books in a similar way, to cover other aspects of mathematics in the same way. For example, optimal estimation, and Kalman filtering, and others.Kudos Nathan, good job.

The book arrived right at my doorstep before I was even expecting it! It has been tremendously helpful to have this book to go back and review material not thoroughly covered in class on MATLAB. I would highly recommend this textbook to those interested in MATLAB or who are enrolled in a university class.

This book really helped me span the bridge between computer science and higher math. There's a lot of material here and it pairs well with the author's online lectures, adding value rather than simply repeating.I've yet to cover partial differential equations so at times found this challenging but otherwise found it approachable for someone without math background.

This is an unusually interesting textbook and professional reference, full of innovative examples of important mathematical modelling techniques. It is the textbook for the excellent free graduate-level MOOC courses offered by the author and the University of Washington through Coursera.

The front cover is blemished but I'm going to make a mess of this book by the time I finish it.I would have bought used if I could.