With the rapid development of technology and the rapid growth of data, data science has begun to penetrate all walks of life. Data science is an interdisciplinary field. It is not easy to get started. The content that needs to be learned includes SQL, Python programming language, web crawlers, data analysis, machine learning, data mining, and data structure algorithms. The recommended reading list for introductory data science is compiled below. Interested friends can take a look. I believe it will be helpful to your study.
Dark horse programmers-a must-read list for big data subjects
1. SQL book list recommendation:
"SQL Study Guide": SQL is a must to learn data analysis and mining. This book comprehensively and systematically introduces the basic knowledge of all aspects of SQL language and some advanced features, including SQL data language, SQL program language, data set operations, subqueries, built-in functions and conditional logic. Readers can quickly master the SQL language by learning this book step by step, and can also read this book to solve related problems when they encounter problems in actual work.
2. Python book list recommendation:
"Think Python Like a Computer Scientist": This book teaches Python programming in accordance with the idea of cultivating readers to think like a computer scientist. The main body throughout the book is how to think, design, and develop methods, and the specific programming language only provides a medium for convenient introduction of specific scenarios. And the book starts from a beginner's point of view, using vivid examples and rich exercises to guide readers into a better environment.
"Head First Python": Go beyond the boring grammar and method manual, teach you to learn this language through a unique method. You will quickly master the basics of Python, and then move on to persistent storage, exception handling, web development, SQLite, data processing, and Google App Engine. A lot of illustrations in the book use Python to solve some practical problems. Beginners can not type the code according to the content of this book, just understand it. The shortcomings are also more obvious, because the amount of code in this book is too large, which may make beginners get started to give up.
"Basics of Python Programming": This book is recommended to friends who like exam-oriented education. It is also very suitable as an introductory book. It is written based on Python3. Besides, Python has been added to the National Computer Second-Level Examination. This book is also very OK as a textbook .
"Smooth Python": This book is dedicated to those who want to write more elegant Python. It analyzes the in-depth content of Python and reads it carefully. Each chapter has a great harvest. Because the book is still very thick, it is even more important for everyone to read it carefully.
3. Recommended web crawler book list:
"Proficient in Python Web Crawlers": I learned web crawlers from the video of Teacher Wei Wei, which is easy to understand and great! After the publication of Teacher Wei Wei’s book, I read it again. The book explains the principles of web crawlers, urllib library, regular expressions, and scrapy more thoroughly. I recommend that you also take a look at the powerful parsing libraries such as BeautifulSoup and xpath. Use Fiddler to capture packets for analysis, but I prefer the developer mode of Google Chrome.
Recommended data analysis book list:
"Introduction to Data Analysis": This book is suitable as an introductory book for learning data analysis. The book contains a large number of illustrations, vivid images, and simple explanation. Each chapter is repeated thinking and iteration to solve specific problems. It is highly recommended to beginners who have no basic knowledge.
4. Machine learning book list recommendation:
"Vernacular Big Data and Machine Learning": I recommend this book for introductory machine learning. Don't entangle the code in the book. Look at the examples of each type of machine learning. It is easy to understand. There are many comic illustrations in the book. The algorithm principle is basically not deep. Especially the Hidden Markov part is great, as an introductory machine learning book is great!
"Machine Learning": This is a classic book that you must read to get started with machine learning. This book starts with the watermelon data and ends with the watermelon data, and ends at each algorithm point (does not mean that the depth is not enough, here is no nonsense in space). Especially the model evaluation selection in Chapter 2 of this book is very systematic. Of course, the theoretical derivation in the book is difficult. If you are interested, you can take a look at some study notes.
"Machine Learning Basic Course": This book is a theoretical book. Starting from linear regression, the least squares method and the maximum likelihood method are all derived in detail, heartily, and then the Bayesian method and Bayesian inference, Although the theory is very strong, it is easy to understand, and the subsequent classification and clustering dimensionality reduction are not featured. Students who are good at math can take a challenge.
"Machine Learning in Action": This book is very powerful. Basically, you don’t need wheels made by others such as sklearn. Basically, you define your own functions to implement functions. It is very helpful for you to understand the implementation of machine learning code from the bottom. There are not enough comments. You may If you don’t understand the meaning of a certain piece of code, you can use the print function to output it, and then understand it. If you encounter a method you haven’t seen before, you can use Baidu’s function. Because the book is relatively old, some methods have been changed or not used. Try Baidu's latest alternative method.
5. Data mining book list recommendation:
"Introduction to Data Mining": It is an introduction. Don't think that the book is easy to understand. It is still difficult. The first chapter is an introduction, and the second chapter discusses the concept of data in detail. You will come into contact with a lot of technical terms that have not been heard. There are not many classification algorithms. The introduction of kernel functions in SVM is very vivid. A lot of pen and ink have been digging deep in the two areas of association analysis and cluster analysis. At the end of this book, we talked about anomaly detection. The book is full of theory, not code implementation.
6. Data structure algorithm book list recommendation:
"Dahua Data Structure": If you are still a student at school and want to recruit a data analysis mining or machine learning position through the school, you will inevitably face data structure algorithm problems. If you are from a non-major class and are new to data structure algorithms It will definitely be awkward, then I recommend you this book, there are a lot of illustrations in the book, to help understand, class-style teaching, very good, to help you open the door to data structure algorithms, let you easily get started!
The above is the recommended reading list for introductory data science. What do you think? Nowadays, data science has become a hot spot for learning, and many learners are starting to learn with their high salaries and promising prospects. Therefore, everyone must understand that in this industry, what is most needed is talents who continue to learn and enrich themselves. Come on, waiting for everyone is the bright side!