The Perfect Data Science Book for beginners, really!

this video is sponsored by brilliant and i'm going to tell you a little bit more about that later i'd like to share this book with you it's the data science design manual by stephen skinner now steven skeener is a professor of computer science at stony brook university he's also written quite a popular book on algorithms which you may be familiar with i don't see this book come up very often in lists of data science books and i think it's quite an omission because this is an excellent book on data science and if data science is something that you want to learn i think you should take a look at this book because it may be very useful to you indeed it covers the data science landscape and tells you what you need to know in order to be able to do data science it draws on stephen skinner's experience as a well he's a computer scientist and a data scientist and so there are anecdotes in there about why you should do things in a certain way the chapters have lots of questions at the end of each chapter there are exercises to work through which really help with understanding and it it helps you find data sets and work through data sets and it asks some questions that really make you think and gain understanding of the topic aren't i always saying the way to learn something is to do it and that's why i want to introduce you to today's sponsor brilliant now brilliant is an interactive online stem learning platform where you can learn you know all the stuff that you're interested in so computer science and data science that's programming and algorithms and data structures and statistics and probability but you learn it interactively and i think that's what makes brilliant unique i've been using brilliant for years long before they reached out to me to sponsor the channel so thank you for getting in touch brilliant what i really like about brilliant is the interactive problem solving because it really ensures that you understand the subject because you've been asked questions about it and you've had to apply that knowledge to different scenarios brilliant has a huge catalogue of courses and it's constantly expanding so there's bound to be something there for you it's particularly strong in data science and computer science so go and take a look at those to get started go to brilliant.org forward slash python programmer or just click on the link in the description and the first 200 people to sign up will also get 20 off brilliant's annual premium subscription so let's take a look at the contents and you'll see what i mean so what's the aim of this book well steven skinner writes in the preface here i stress the following basic principles is fundamental to becoming a good data scientist valuing doing the simple things right developing mathematical intuition and to think like a computer scientist but act like a statistician so at the beginning he says data science isn't rocket science students and practitioners often get lost in technological space however the heart of data science lies in doing the simple things right understanding the application domain cleaning and integrating relevant data sources and presenting your results clearly to others and so what does it cover well to start with it covers a sort of high-level look at what data science is it then goes on to cover the mathematical preliminaries so this is the math that you're going to need to know in order to be able to do data science then data munching that's cleaning data so how do you combine different data sets and and distill them down to something that you can work with and you know use data science methods on scores and rankings statistical analysis visualizing data where else mathematical models linear algebra linear and logistic regression distance and network methods machine learning and big data so these are all topics that usually are in a lot of different books you know so when you're learning data science you need to learn linear algebra you go off and get a linear algebra book but of course linear algebra is a vast field and really as a data scientist you know there are certain bits that are more uh that are more relevant to what you're doing and the same is true with all of these other topics the book covers what this book aims to do is to cover all of the topics that you need as a data scientist in enough detail for you to be able to go off and do data science let's have a look at the chapter on statistical analysis so here it is statistical distributions so we cover the binomial distribution the normal distribution uh the poisson distribution power law distributions sampling from distribution so obviously sampling is a really important aspect of data science and something that you need to understand statistical significance permutation tests and p values and bayesian reasoning so let's look at one of the questions for example explain which distribution seems most appropriate for the following phenomenon binomial normal poisson or power law the number of leaves on a fully grown oak tree the age at which people's hair turns gray the number of hairs on the heads of 20 year olds the interview questions are interesting as well what is conditional probability what is bayes theorem and why is it useful in practice and how would you improve a spam detection algorithm that uses a naive bayes classifier i use this book as a reference so if i'm working on a data science problem and there's something that i can't remember or i'm a bit unsure of this is the book that i go to first there are other books that i use as well but this is always the book that i go to first i think it's an excellent book and if data science is something that you're interested in maybe you're doing a masters in data science or you're doing an online data science course then this would certainly be a good companion to that also if you're not studying data science but you're interested in learning data science this would be a fairly inexpensive way of seeing whether the subject is something that you think you could enjoy having said that this book isn't cheap i think i paid about 40 pounds for it but that's certainly cheaper than starting a course that ends up not being right for you and it covers a lot of information all in one place that would take you a lot of time to find and put together yourself there's also a companion website which has a lot of material on it as well and stephen even created a tv series that followed some of his students at stony brook as they were doing their data science masters so you can take a look at that too so all in all i think 40 pounds is pretty good value and i think it would be a useful addition to the library of anyone that wanted to learn data science and it's not one that comes up very often so definitely take a look at it i think it could be really useful for you

Channel: Python Programmer

Views: 47,084

Keywords: learn python, python, data science, learn data science

Id: 2ow93RSqc5g

Length: 7min 1sec (421 seconds)

Published: Tue Mar 29 2022