Data Analysis in Practice

Name: Data Analysis in Practice
ISBN: 9787115454539

Ryuji Sakamaki / Yohei SatoPosts & Telecom Press2017

3.5

Word count: 258K words
Chapters: 12
Series: Turing Programming Series
ISBN: 9787115454539

Report an error Find this book

Summary

This book is co-authored by two frontline data analysts working in the data analysis department of DRECOM in Japan. It is aimed at readers who need to do data analysis in real business settings, not beginners who only want to learn software operations. The book first explains, in Part I, the roles and division of labor among data scientists and the current state of the field, and presents a reusable five-step workflow for business data analysis: grasp the current situation and expectations, identify the problem, collect and process data, perform the analysis, and propose solutions. Parts II and III each center on four real business cases: why sales are declining, what kind of customers churn, which advertising works better, how to acquire more users through ad combinations, as well as using logistic regression to predict user behavior, using principal components and clustering to segment target user groups, using decision trees to identify the behavioral characteristics of long-term users, and using machine learning to make team-based games more fun. Each case follows the same workflow from start to finish, and detailed R code is provided at the end of each chapter so readers can reproduce the analysis. The book deliberately uses raw, uncleaned data, requiring readers to organize it themselves, making the experience closer to real day-to-day analysis scenarios. At about 254 pages, the book is compact, and its focus is not on deriving algorithms but on "how to translate a business problem into an analytical problem, and then translate the analytical results back into business decisions."

Editor's review

市面上讲数据分析的书大致分两类：一类讲工具，一类讲算法，而真正缺的是“怎么从头到尾做完一个分析项目”的书。这本正好补了这一块。它的骨架非常清晰——先把商业问题拆成可验证的假设，再去找数据、洗数据、建模，最后一定要落到“所以要做什么”。每个案例都走完这一整套流程，读者读上两三章就能形成肌肉记忆。另一个可贵之处在于数据是脏的：很多教程为了讲解方便，把数据整理得干干净净，读者回到工位就傻眼；这本书反过来，把清洗和加工的过程完整摊开，甚至包括数据拿不到正解时怎么办。附带 R 代码可下载，边读边敲完全可行。全书只有 254 页，一个周末能通读一遍，之后再当手册按需翻查案例。需要提醒的是，作者刻意不深挖算法原理，如果你要找的是数学推导或调参技巧，会觉得浅；但如果你想建立“业务—数据—决策”的完整链条，它是很好的入门与参照。

Who it's for

A good fit for

刚入行 1—2 年、想把分析做完整而不只是出报表的数据分析师
需要用数据支撑运营决策的产品经理与运营人员
会一点 R 或 Python、但不知如何切入真实业务问题的学生
做增长、广告投放或用户留存，需要 A/B 测试与用户分群实战参照的从业者

Not a good fit for

希望系统学习统计理论与算法数学推导的读者
完全没有编程基础、只想看概念性科普的入门读者
寻找深度学习、大数据工程等前沿方向内容的资深算法工程师

Reading tips

先读第 1、2 章建立流程框架，把五步法记牢，再进入案例部分
每个案例先自己看数据想思路，再对照书中的分析和代码，效果远好于直接照抄
把章末 R 代码跑一遍，并尝试用 Python 重写，检验自己是否真正理解每一步
读完可以挑一个自己工作中的真实问题，按书中五步完整走一遍作为练习

About the publisher

Posts & Telecom Press China · Founded in 1953

Posts & Telecom Press is known for its strengths in science and technology, computing, and information and communications titles, and has extended its publishing into photography, art, software learning, and inspirational growth, giving it a fairly diverse range. Its catalog includes computing and communications works such as Refactoring and Patterns (Revised Edition) and Huawei Switch Learning Guide (Second Edition), alongside bilingual collector's editions such as The Little Prince (Chinese-English-French illustrated collector's edition), The Great Gatsby, and Three Days to See. Beijing Ten Years: 1980-1990 and Ghost Blade: WLOP Personal Illustration Collection reflect its presence in photography and art. Its authors include Antoine de Saint-Exupéry, Helen Keller, Joshua Kerievsky, Wang Da, Zhang Zhaozeng, and WLOP.

Chapter 1 The Work of Data Scientists
Chapter 2 The Business Data Analysis Workflow
Analysis Fundamentals
Chapter 3 Case 1—Bar Charts: Why Are Sales Declining?
Chapter 4 Case 2—Cross-Tabulation: What Kind of Customers Leave?
Chapter 5 Case 3—A/B Testing: Which Advertisement Works Better?
Chapter 6 Case 4—Multiple Regression Analysis: How to Acquire More Users Through Combinations of Advertisements
Analysis Applications
Chapter 7 Case 5—Logistic Regression Analysis: Can the Present Be Predicted from Past Behavior?
Chapter 8 Case 6—Clustering: Which Target User Group Should We Choose?
Chapter 9 Case 7—Decision Tree Analysis: Which Behaviors Identify Long-Term Users?
Chapter 10 Case 8—Machine Learning: How to Make Team-Based Games Fun

FAQ

这本书主要用什么工具做分析？

正文以分析思路为主线，操作层面配套 R 语言代码，每章末尾有“详细的 R 代码”小节，数据和脚本可下载。书中不要求读者精通 R，能照着跑通、看懂输出即可，思路本身也可以用 Python 或 Excel 复现。

没有统计基础能读懂吗？

可以。书中涉及柱状图、交叉列表、回归、聚类、决策树等方法，但只讲“什么时候用、怎么看结果”，不做数学推导。有高中统计常识就能跟上，遇到不熟的概念可配合其他教材补充。

和同名的其他《数据分析实战》是同一本吗？

不是。市面上有多本同名书，例如托马兹·卓巴斯著、机械工业出版社版，纪贺元著、机械工业出版社版，以及汝思恒著、清华大学出版社版。本页指的是日本作者酒卷隆治、里洋平合著，肖峰翻译，人民邮电出版社 2017 年出版、ISBN 9787115454539 的这一本。

书里的案例是什么行业的？

八个案例多取自互联网与游戏服务场景，涉及销售额下滑归因、顾客流失、广告效果评估、广告组合投放、用户行为预测、目标用户分群、长期用户识别和组队玩法优化，方法论可迁移到其他行业。

Comments

Find this book

This site is an index: you start “Search all sources” in your browser, and the entries it lists all point to third-party sites; please judge their availability yourself.