Sharing today: Which books to read for introductory data analysis
The editor visited Zhihu and websites related to data analysis today, and selected ten highly searched books, and I recommend them to everyone here. Regarding the level of data analysis ability, mainly according to the following three stages:
Getting started
The introductory chapter recommends four books, suitable for those who are just beginning data analysis: you who want to change careers, you just graduated, and you who have no overall concept of data analysis, you can start to read these books first.
1. The
main purpose of "Pyramid Principle" is to cultivate and establish a personal logical thinking system and expression.
The principle of the pyramid emphasizes: the key points, clear logic, clear primary and secondary logical thinking; the basic structure of the pyramid is: clear central ideas, conclusions first, unify the above, categorize and group, logically progressive, first important then secondary, first overall After details, first conclusions and then reasons, first results and then process; Pyramid trains the expresser: pay attention to and explore the audience's intentions, needs, interests, points of interest, and excitement, figure out what the content says, how to say, and master the expression The standard structure and standardized actions of the pyramid; the pyramid helps to achieve the purpose of communication: the focus is prominent, the thinking is clear, the primary and the secondary are distinct, so that the audience is interested, understandable, acceptable, and remembered; the specific method of building the pyramid is: top-down expression, Bottom-up thinking, vertical question answering/summary summary, horizontal classification and grouping/deduction induction, preface tells the story, title extracts the essence of thought.
2. "In-depth explanation of data analysis"
This book vividly shows readers the techniques that excellent data analysts should know and should know (basic steps of data analysis, experimental methods, optimization methods, hypothesis testing methods, Bayesian statistical methods) in a form similar to "chapter-back novels". , Subjective probability method, heuristic method, histogram method, regression method, error processing, related databases, data collation techniques).
3. "Who said a rookie can't analyze data"
This book has made great attempts to reduce the difficulty of learning: it mainly uses the familiar Excel tools, plus some necessary data analysis concepts, and uses easy-to-understand explanations to make the content easy to understand and intangible. I learned data analysis in China. The book consists of 8 chapters, respectively explaining the necessary knowledge of data analysis, data processing skills, data presentation technology, improving the beauty of charts through a professional perspective, writing skills of data analysis reports, and continuous training.
4. "Naked Statistics"
Whenever I hear statistics, I feel a headache to some extent, but this book completely avoids statistics (how is it possible, kidding), the point is: this book does not have mathematical formulas that you can't avoid, it is not full It is a digital chart, no boring textbook preaching; the book is interspersed with some lively and humorous cases, such as some social problems, life problems, and use statistical knowledge to analyze and explain them.
Advanced
These books have a certain industry specificity (but the basic principles are general), and readers need to have a certain analysis knowledge, suitable for website analysts, business analysts, and data product managers who have data analysis experience.
1. "Proficient in Web Analytics 2.0"
In the past few years, the Internet, online marketing, and advertising have undergone tremendous changes. However, the way people deal with data is still much the same as it was a few decades ago. It is stagnant, how to transform to data-driven decision-making and how to use website data. To gain a competitive advantage, this has become the focus of Internet companies.
In this book, the author proposes a framework for next-generation website analysis, and explains how to improve the organization's initiative and response speed to the market. Through the transformation of the traditional method, the author analyzes the Internet data like a whistle, and proposes a specific, simple and more advanced method. If you want to become a website analysis expert, this book will be your best choice.
2. "MySQL Must Know and Know"
The book starts from the introduction of simple data retrieval, and gradually goes deep into some complex content, including the use of joins, subqueries, regular expressions and full-text-based searches, stored procedures, cursors, triggers, table constraints, and so on. Through the highlighted chapters, the knowledge that readers should be mastered is described in a clear, systematic and concise manner, so that your data skills will be greatly increased inadvertently.
3. "Using python for data analysis"
This book contains a large number of practical cases. You will learn how to use various Python libraries (including NumPy, pandas, matplotlib, and IPython, etc.) to efficiently solve various data analysis problems.
Advanced articles
Higher-level data analysis is relatively more professional, such as corporate internal data governance, business analysis of data integration, and data visualization. Of course, there are more in-depth things like data mining algorithms. You are suitable for a big cow or you who want to become a big cow, provided that your data analysis ability has reached this stage.
1. "Fighting Big Data"
This book puts perspectives into the field of "big data practice", and it is important for big data applications such as data collection, data operation, operational data, wireless data, data blind spots and noise, data classification and data value, data maintenance, and multi-screen era. He made detailed answers to hot issues, thought about today’s big data from multiple perspectives, and put forward forward-looking suggestions for “personal big data management”, and created a closed-loop system for data-based operations and operational data. At the same time, "Fighting Big Data" unveiled the mystery of Alibaba's operational data for the first time, deciphering its data practice's "mixing, communicating, and exposing" internal three axes and "storing, managing, and using" external three axes. It is very meaningful for most e-commerce companies.
2. "The Beauty of Data"
This book uses some examples of data workers to show readers how to process data. This allows readers to stand on the shoulders of excellent data designers, managers, and processors to carefully examine some of the most interesting projects involving data.
3. "Data Science in Action"
This book should be a bridge between data analysis and data mining (machine learning). Starting from exploratory data analysis, through the thinking of data analysis, the basic algorithms of machine learning are derived: regression analysis, k nearest neighbors, and k means. Then the most common machine learning algorithms are introduced through different application scenarios, as well as their applications in real scenarios. For those who have done data analysis for a period of time, this is undoubtedly a good book for advanced and higher dimensions. It is difficult to have a book that can smoothly transition you from simple data analysis to machine learning and data mining, so if you After doing some exploratory analysis and encountering a bottleneck, you will naturally enter the pit of data mining and machine learning algorithms, because only more advanced algorithms and models can support large-scale data prediction, then you can take a look at this book Book it.