# A Comprehensive Review of The Data Science Design Manual

Written on

## Chapter 1: Overview of The Data Science Design Manual

The book "The Data Science Design Manual" by Steven S. Skiena spans 462 pages and was published by Springer on July 1, 2017. It’s important to note that the Kindle version suffers from significant formatting issues, particularly with mathematical expressions.

### What Can You Expect from This Book?

Within its pages, you’ll find:

- Essential mathematics for data science (and there’s a lot of it!)
- A spectrum of algorithms, from basic to advanced
- Statistical methods and analyses
- An extensive collection of exercises and homework tasks
- An introduction to machine learning models
- Access to the Quant-Shop, featuring video examples with accompanying code

While I appreciate the content in "The Data Science Design Manual," I must clarify that my critique isn’t an outright condemnation of the book. However, I do question its intended audience. The back cover suggests it is designed for undergraduate or postgraduate data science students, but I find it may be overwhelming for the former group and insufficient for the latter.

If you delve deeper, you’ll discover that Skiena has authored two other notable works:

- "Calculated Bets: Computers, Gambling, and Mathematical Modeling to Win"
- "Who's Bigger?: Where Historical Figures Really Rank"

He frequently references these books, along with his Quant-Shop, throughout the text. While this could be beneficial, his focus on theoretical concepts rather than practical applications can be quite draining. Although he includes real-world examples, they often relate back to his previous works, which can lead to repetition.

### The Exercises: A Strong Point

One of the elements I found most valuable was the extensive exercises, homework assignments, and Kaggle challenges. These require a commitment of time that often exceeds the time spent reading the corresponding chapters.

However, a significant drawback for me was the Kindle version’s formatting, which renders mathematical formulas, tables, and other expressions nearly unreadable. Though none of the formats are inexpensive, I would hesitate to recommend purchasing this book in any form due to these issues. While the content does hold merit, much of the learning material is available at a lower cost elsewhere.

Skiena's writing style tends to be dry and math-focused, often leaving the reader longing for deeper insights. Notably, there are no specific coding examples; instead, he instructs readers to create models in their preferred programming language, which might be daunting for novices.

### Potential Audience and Value

As for who this book is best suited for, I’m at a loss. Perhaps it appeals to those with a significant budget who are willing to invest in the hardcover or paperback editions, priced at $75 and $65, respectively. Ultimately, I see this book as a decent reference but not much more.

The Quant-Shop videos, which were marketed as a selling point, are of relatively low quality and may not meet expectations for modern educational content. While they may have been valuable in 2017, they now seem outdated, lacking the polish expected in today’s resources.

In conclusion, while I appreciate "The Data Science Design Manual," I find its pricing and Kindle formatting frustrating. There are superior, often free resources available that cover similar material.

★★★★☆ 4/5 stars

## Chapter 2: Supplementary Learning through Video Resources

To supplement your understanding of data science concepts, consider watching the following videos:

This comprehensive tutorial covers Natural Language Processing (NLP) in Python, complete with practical examples that enhance your learning experience.

The "Python For Everybody" course provides solutions to exercises from Coursera and edX, making it an excellent resource for both beginners and experienced learners.