How to Learn DATA SCIENCE in 12 Months – ‘Course Review’

We all know that Data Science is one of the hottest skills in 2017. It’s also very demanding in IT industry. The demand for data scientist has exponentially increased in last couple of years and it doesn’t seem to slow down anytime soon. As per IBM, by 2020, around 1.7 million data-related job openings are expected to be created in the U.S.

This is the right time to learn data science if you still didn’t started it. If you don’t know where to start, this article will help you to take first step for learning Data Science. I’m currently taking Andrew Ng’s course on Coursera – “Machine Learning “. It’s a great course, but also very time-consuming. But I know how it feels when you don’t know where to start and how to learn this complex subject.

So, I got idea of writing an article based on my experience of learning Data Science so that it will help other people as well.  

This article is not for advanced learners or experienced data scientists who already know most of the stuffs covered in this course. This article is meant for beginners only who just started their journey of learning Data Science and looking for useful resources. Please note that all the time duration mentioned here is approximate and may vary depending upon your skill and effort level on respective topic. 

I’m taking this course from August 2016 and right now I’m in my 10th month. So, to me it’s a 1 year course so this article will be helpful for people who want to take 1 year or one semester off from their regular work and dedicate full time for learning Data Science. You can take the assistance of the DBA administrator.

Here is the list of all the topics that I’ll cover in this article:

Good luck with your machine learning journey! Please feel free to request topics you want to be covered here. –> [Link]

#1> Basic Statistics – Time Required : 3 months (+3 weeks)  Difficulty : Easy Breakdown: 23 Days Intro To Statistics 7 Days Descriptive Statistics 5 days Inferential Statistics 8 Days Regression Analysis 9 Days Classification Analysis 4 Days Further Study On Machine Learning In R In last few years, Machine Learning has become an integral part of not only Computer Science but also other fields like Biology, Astronomy etc. It is a branch of Artificial Intelligence that gives computers ability to learn without being explicitly programmed. For example – We all know Google’s PageRank algorithm right? But do you know the algorithm learns from its user behavior and search queries they type on Google Search Engine and accordingly updates the page ranking of websites every month or so? To understand this example better, let me explain what is Machine learning briefly…. In simple words “Machine Learning” is a way to teach computer programs to make predictions based upon existing data rather than explicit programming. So basically machine learning algorithms give computer systems the ability to automatically learn and improve from experience without being explicitly programmed. Before going further, let’s talk about some basic terminologies first…

Terminology: 

Algorithm: 

A specific set of unambiguous instructions for a computer to perform a task Learning: Acquiring knowledge or skills through study, instruction, or experience Prediction/Forecast: To determine or calculate something in advance based upon available information Example – In Google’s PageRank algorithm, the web pages are being ranked using machine learning algorithm that continuously learns from user search queries and internet traffic of each page. #2> Python For Data Science – Time Required: 12 weeks (+6 days) Difficulty: Medium Breakdown: 70 Days Intro To Python 8 Days Numpy Array 9 Days List & Dict 5 days Pandas Data Frame 9 Days Matplotlib 5 days Scikit-Learn Machine Learning Algorithms 9 Days Statistics In Python: StatsModels, Seaborn Another important topic that is very closely related to machine learning and is a must for deep understanding of the subject is knowing how to program in Python. Its advantage is it has vast collection of scientific computing libraries like Numpy, Pandas etc which makes it very popular among data scientist community as well as Kaggle competitors. I’m not saying you should know all those libraries but at least you should know basic constructs such as List & Dictionary, Numpy Array because almost all Machine Learning packages are built upon this library.

Conclusion:

So to summarize, basic statistics can be covered in 3 months whereas Python for Data Science takes 6 more months. Regression, Classification Analysis is further topics that you should cover after this.

Leave a Comment