|
||||||||||||||||||||
|
||||||||||||||||||||
|
||||||||||||||||||||
よく一緒に購入されている商品
商品の説明
内容説明
Collecting data is relatively easy, but turning raw information into something useful requires that you know how to extract precisely what you need. With this insightful book, intermediate to experienced programmers interested in data analysis will learn techniques for working with data in a business environment. You'll learn how to look at data to discover what it contains, how to capture those ideas in conceptual models, and then feed your understanding back into the organization through business plans, metrics dashboards, and other applications. Along the way, you'll experiment with concepts through hands-on workshops at the end of each chapter. Above all, you'll learn how to think about the results you want to achieve -- rather than rely on tools to think for you. * Use graphics to describe data with one, two, or dozens of variables * Develop conceptual models using back-of-the-envelope calculations, as well as scaling and probability arguments * Mine data with computationally intensive methods such as simulation and clustering * Make your conclusions understandable through reports, dashboards, and other metrics programs * Understand financial calculations, including the time-value of money * Use dimensionality reduction techniques or predictive analytics to conquer challenging data analysis situations * Become familiar with different open source programming environments for data analysis "Finally, a concise reference for understanding how to conquer piles of data." --Austin King, Senior Web Developer, Mozilla "An indispensable text for aspiring data scientists." --Michael E. Driscoll, CEO/Founder, Dataspora
著者について
Philipp K. Janert is Chief Consultant at Principal Value, LLC. He has worked for small start-ups and in large corporate environments, both in the US and overseas, including several years at Amazon.com, where he initiated and led several projects to improve Amazon's order fulfillment processes. Philipp K. Janert has written about software and software development for the O'Reilly Network, IBM developerWorks, IEEE Software, and Linux Magazine. He holds a Ph.D. in Theoretical Physics from the University of Washington. Visit his website at www.principal-value.com.
登録情報
|
この商品を見た後に買っているのは？
類似した商品から提示されたタグ(詳細)関連タグ(この商品に近い関連キーワード)を追加する++最初のタグになります
|
カスタマーレビュー
Amazon.com で最も参考になったカスタマーレビュー (beta)
Amazon.com:
25件のカスタマーレビュー
145 人中、131人の方が、｢このレビューが参考になった｣と投票しています。
It falls short of initial expectations
2011/2/8
形式:ペーパーバック
This book is aimed at offering a practical, hands-on introduction to data analysis for pragmatic readers without strong scientific or statistical background. Some basic programming experience is required. The author provides many personal (and sometimes useful) comments about different tools and procedures in data analysis.
However, a careful reading reveals many problems, specially an obscure presentation of key concepts. In my opinion, the target audience for this book would be people without previous contact with data analysis. Hence the importance of presenting its core elements correctly. Otherwise, it's useless for them.
In particular:
- Few pages are actually dedicated to present open source tools supporting the different graphs and techniques included in the book. From the title, I expected a more complete tour through available open source tools for data analysis.
- No clues about how to obtain most of the graphs and results presented in the book. No related data sets are available for download, either. A book like this is useless if we cannot learn how to replicate all the examples.
- The formula of the variance for a sample is just wrong. One must divide by n-1 and not n; see "Applied Statistics and Probability for Engineers" (Montgomery and Runger 2006).
- The author presents one of the most obscure explanations for the median I've ever come across. Recurring to an RFC (RFC 2330) to explain such a simple concept is really awkward.
- In chapter 3 and Appendix B, natural logarithms (base e) are presented in the text, while graphs plot powers of 10. Definitely, not the right way to transmit correct concepts and methods.
- I concur with a previous review in that "Workshop" sections just present an ultra-short overview of some open source tools. A quick search in your favourite engine will display much more informative introductions (even quick start guides).
- Today, effective data analysis heavily depends on using the best possible implementation. While I might find educational to learn some of this implementations, in a real situation it is much better to rely on precise implementations of algorithms already available (e.g. libraries in GNU R).
All in all, I still recommend "R in a Nutshell" for a gentle introduction to data analysis with an open source tool (GNU R). It also has some inaccuracies and typos, but at least it's much more informative and clear. Besides, it does include an R package with all datasets and examples, ready to be installed and explored.
However, a careful reading reveals many problems, specially an obscure presentation of key concepts. In my opinion, the target audience for this book would be people without previous contact with data analysis. Hence the importance of presenting its core elements correctly. Otherwise, it's useless for them.
In particular:
- Few pages are actually dedicated to present open source tools supporting the different graphs and techniques included in the book. From the title, I expected a more complete tour through available open source tools for data analysis.
- No clues about how to obtain most of the graphs and results presented in the book. No related data sets are available for download, either. A book like this is useless if we cannot learn how to replicate all the examples.
- The formula of the variance for a sample is just wrong. One must divide by n-1 and not n; see "Applied Statistics and Probability for Engineers" (Montgomery and Runger 2006).
- The author presents one of the most obscure explanations for the median I've ever come across. Recurring to an RFC (RFC 2330) to explain such a simple concept is really awkward.
- In chapter 3 and Appendix B, natural logarithms (base e) are presented in the text, while graphs plot powers of 10. Definitely, not the right way to transmit correct concepts and methods.
- I concur with a previous review in that "Workshop" sections just present an ultra-short overview of some open source tools. A quick search in your favourite engine will display much more informative introductions (even quick start guides).
- Today, effective data analysis heavily depends on using the best possible implementation. While I might find educational to learn some of this implementations, in a real situation it is much better to rely on precise implementations of algorithms already available (e.g. libraries in GNU R).
All in all, I still recommend "R in a Nutshell" for a gentle introduction to data analysis with an open source tool (GNU R). It also has some inaccuracies and typos, but at least it's much more informative and clear. Besides, it does include an R package with all datasets and examples, ready to be installed and explored.
24 人中、24人の方が、｢このレビューが参考になった｣と投票しています。
Full of insight, light on details
2011/4/17
形式:ペーパーバック
This book covers such a wide range of topics that it necessarily skims over all of them but it always hits all the major points that an introductory survey should. Each chapter has a straight forward tone, strikes the right balance between developing mathematical rigor and developing an intuitive understanding of data , and undeniably passes on the lessons of hard earned, real world experience. But a reader who is actually working on a real data problem will almost certainly come to the realization that the understanding gained is somewhat superficial - that it's going to take a lot more heavy reading (probably of books, papers, and software tools recommended in this book) to get any real work done!
The single biggest problem with this book is its misleading title. This book is not going to teach you how to use open source software to analyze data. There is only minimal information about how one would actually use the software tools being discussed. What you get is a brief commentary about what the author thinks each software package is good for. It's the same story as with the mathematical details: you will not find them here, but this book will give you an excellent idea of what to look for. So in the end it does leave you feeling just a little bit cheated, even though all the advice you got seems extremely well informed.
What this book does astonishingly well is communicate an attitude to data analysis that most textbooks (and nearly all the college courses I took) seem to miss. Nearly every chapter is a stream of stunningly insightful observations on how to approach data, without the mathematical detail that overwhelms most practicing programmers. I would recommend it to any reader who understands that truly useful insights are hard to come by, but detailed algorithms and formulae are easily found in the Internet Age. I wish the book were a few hundred pages shorter, that it corrected a few sloppy mistakes (like confusing revenue and profit), but I'm certainly glad I read it.
The single biggest problem with this book is its misleading title. This book is not going to teach you how to use open source software to analyze data. There is only minimal information about how one would actually use the software tools being discussed. What you get is a brief commentary about what the author thinks each software package is good for. It's the same story as with the mathematical details: you will not find them here, but this book will give you an excellent idea of what to look for. So in the end it does leave you feeling just a little bit cheated, even though all the advice you got seems extremely well informed.
What this book does astonishingly well is communicate an attitude to data analysis that most textbooks (and nearly all the college courses I took) seem to miss. Nearly every chapter is a stream of stunningly insightful observations on how to approach data, without the mathematical detail that overwhelms most practicing programmers. I would recommend it to any reader who understands that truly useful insights are hard to come by, but detailed algorithms and formulae are easily found in the Internet Age. I wish the book were a few hundred pages shorter, that it corrected a few sloppy mistakes (like confusing revenue and profit), but I'm certainly glad I read it.
36 人中、34人の方が、｢このレビューが参考になった｣と投票しています。
Good, not great. Prerequisites and chapter organization issues.
2011/1/28
形式:ペーパーバック
The book is very good for the intermediate-to-advanced data analysts. Beginners beware: there are some important prerequisites that are not obvious before you buy it, and there are some organization problems.
First, the prerequisites. "I strongly recommend that you make it a habit to avoid all statistical language"..."Once we start talking about standard deviations, the clarity is gone." These are two sentences in the same passage from the Preface. The rest of that passage is similar. However, even the first chapters make heavy use of statistical language. Moreover, they assume that you already know statistics to the level of density estimation, noise, splines, and regression. Page 21 even features a footnote about the Fourier transform and Fourier convolution theorem. Clearly this book is not for the statistically-shy or for mathematically-shy in general, no matter what the Preface suggests. You also need to know Python and R.
Second, the chapter organization problems. There's a mismatch between the first part of each chapter, which introduces concepts and techniques, and the Workshop part of the same chapter, which uses software. I was expecting the Workshop to illustrate the implementation of the same concepts and techniques. It's not really so. The Workshop introduces Python and R facilities at a different (lower) speed than the rest of the chapter. One could even wonder why the Workshop is in the same chapter. I'd rather that each chapter consisted of a few detailed case studies that first introduce concepts and techniques and then illustrate them with software libraries.
First, the prerequisites. "I strongly recommend that you make it a habit to avoid all statistical language"..."Once we start talking about standard deviations, the clarity is gone." These are two sentences in the same passage from the Preface. The rest of that passage is similar. However, even the first chapters make heavy use of statistical language. Moreover, they assume that you already know statistics to the level of density estimation, noise, splines, and regression. Page 21 even features a footnote about the Fourier transform and Fourier convolution theorem. Clearly this book is not for the statistically-shy or for mathematically-shy in general, no matter what the Preface suggests. You also need to know Python and R.
Second, the chapter organization problems. There's a mismatch between the first part of each chapter, which introduces concepts and techniques, and the Workshop part of the same chapter, which uses software. I was expecting the Workshop to illustrate the implementation of the same concepts and techniques. It's not really so. The Workshop introduces Python and R facilities at a different (lower) speed than the rest of the chapter. One could even wonder why the Workshop is in the same chapter. I'd rather that each chapter consisted of a few detailed case studies that first introduce concepts and techniques and then illustrate them with software libraries.
これらのレビューは参考になりましたか？
ご意見はクチコミでお聞かせください。
リストマニア
|
|
関連商品を探す
同じキーワードの商品を探す
フィードバック
- カタログ情報、または画像について報告
- 著者からのコメント
- 出版社からのコメント
|
