Roadmap in Data Science for Beginners | Coursera Community

Roadmap in Data Science for Beginners

  • 22 January 2021
  • 15 replies
  • 2106 views

Userlevel 3
Badge +1

Hey There,

I am new Data Science Enthusiast who is interested in the this field. But have no clear idea about the road map of Data Science. I was Studying and Working in Web Development before, but now i am changing my career path.

So, it would be great if someone can guide me through.

Regards

-Tejal


15 replies

Userlevel 3
Badge +1

Hi Tejal,

I am in the same boat as you. I also want to become a Data Scientist “When I  grow up!”

There are several places on the web that are really friendly and helpful.

  1. [Dev.to](https://dev.to/) has thousands of tips and hints and articles on all sorts of computer science issues. I believe it started as a place for Javascript web development but it has everything. Check it out.
  2. [Free Code Camp](https://www.freecodecamp.org/) is another really positive and helpful space on the web. They are growing their Data Science course selection too!

Best of Luck,

Let me know what you think of those, I have more… ;))

Userlevel 1
Badge

Hello, which language you’d like to know first? Python or R? 

Userlevel 3
Badge

Hi Tejal

 

1st> learn python (along with libraries)or R


2nd> get idea about  SQL and M.soft XL

 

3rd>  get some basic knowledge about linear algebra , PROBABILITY AND STATISTICS


4th> make your hand dirty with data, start practice, solve problems (ex> KAGGLE)

After that>
5th> Gradually u will understand what u need to do next……….:wink:
 
u can learn  few courses if u wish .I recommend (python for everybody by Dr. Charles Severance)  really a nice course for beginner...:muscle:

Best of luck….:thumbsup:

Userlevel 3
Badge +1

Hi All,

For the most part, I agree with Jessun.

Really R and Python are the most widely recognized languages for Data Science. Languages are like candy, everyone has their favorite. Both have their strengths and weaknesses. Learning one will always help you build on another.

I will say this; It is better to learn one language REALLY well than learn “Hello World” in ten different languages.

There are so many resources out there. In the beginning, I did not pay for anything. I used free sites to learn. But after a while when you want to focus or have specific questions then you can look for (buy) a book that will help.

Another thing I will add is; Don’t shy away from new or ‘hard’ items. Growing/Learning means trying new and different things. Go at your own pace and don’t burn out.

Good Luck:grinning:

Userlevel 1
Badge

1st> learn python (along with libraries)or R

2nd> get idea about  SQL and M.soft XL

3rd>  get some basic knowledge about linear algebra , PROBABILITY AND STATISTICS

4th> make your hand dirty with data, start practice, solve problems (ex> KAGGLE)

After that>
5th> Gradually u will understand what u need to do next……….:wink:
 

I generally agree with this, but I’ll say I think it’s better to do a bit of 1 and 3 at the same time.  It’s hard to understand entirely what you’re learning in Python or R about doing statistical things unless you get some of the statistical concepts being applied there.

 

Userlevel 2
Badge

Piggybacking off Tejal’s list of priorities, I’ve followed this basic path and have taken the following courses here in Coursera. This has helped prepare me for applying to Master’s of Analytics program at Georgia Tech for this coming Fall.

  1. Learn Python basics, data structures, and intermediate with functions - Python 3 Programming Specialization [5 courses - last is a final project]
  2. Then, put your Python skills to the test and apply what you’ve learned to dig deeper into Data Science skills in Python - Applied Data Science with Python Specialization [5 courses]
  3. Great Introductory SQL and database course. Start learning SQL from the ground up and apply what you learn to data analysis in SQL and databases. Learn SQL Basics for Data Science Specialization [4 courses - last is a capstone project]
  4. I’m getting ready to take Math for Machine Learning Specialization to review (1) Linear Algebra, (2) Multivariate Calculus, and (3) Principal Component Analysis (PCA) all centered around its use in Data Science and modeling.

I’ve enjoyed these tremendously so far.

Enjoy,

Brannon

Userlevel 1
Badge +1

Piggybacking off Tejal’s list of priorities, I’ve followed this basic path and have taken the following courses here in Coursera. This has helped prepare me for applying to Master’s of Analytics program at Georgia Tech for this coming Fall.

  1. Learn Python basics, data structures, and intermediate with functions - Python 3 Programming Specialization [5 courses - last is a final project]
  2. Then, put your Python skills to the test and apply what you’ve learned to dig deeper into Data Science skills in Python - Applied Data Science with Python Specialization [5 courses]
  3. Great Introductory SQL and database course. Start learning SQL from the ground up and apply what you learn to data analysis in SQL and databases. Learn SQL Basics for Data Science Specialization [4 courses - last is a capstone project]
  4. I’m getting ready to take Math for Machine Learning Specialization to review (1) Linear Algebra, (2) Multivariate Calculus, and (3) Principal Component Analysis (PCA) all centered around its use in Data Science and modeling.

I’ve enjoyed these tremendously so far.

Enjoy,

Brannon

Hi, Brannon

 

Thank you so much for your suggestion. I am currently taking a course in Visualization for Data Journalism. And I was struggling with the Data Wrangling thing. I learned Python 3 from Kaggle and now, I just enrolled in the Python 3 Programming Specialization as you suggested. After this, I am going to enroll in Applied Data Science. 

 

Warm Regards, 

Karla

Badge

Thanks everybody for your suggestions.

We could also talk about books. I recommend reading “The Master Algorithm” by Pedro Domingos. You can find a good summary of the main schools of machine learning (apart from the view of the author about these issues and his proposals for a “master algorithm”). It’s quite technical and sometimes you get lost but mostly you can get the basic ideas.

It is a well written book, with lots of colorful examples. As the author seems to be a real polymath, you can find many references to diverse issues like philosophy, science or psychology. Reading this book takes times but it is worth the effort.

I’m also very interested in reading Nate Silver’s “The Signal and the Noise”, although I have not started it yet. 

 

Badge

Thanks everybody foor your suggestions

 

Userlevel 4
Badge +7

I would suggest following this path -

  1. I suggest learning Python and then linear algebra/data science/ml libraries in Python. Eg. Numpy, Pandas and Scikit learn. They are pretty easy to learn once you have mastered Python. I would suggest using one visualization library like Plotly or Seaborn or Matplotlib.
  2. You should find courses on Coursera and elsewhere to learn this.
  3. This and doing some projects available on Kaggle, on the internet should help you get practice.
  4. For advanced machine learning and deep learning, I would suggest reviewing the math being deep learning and then choosing a Python library like Keras(high-level Tensorflow wrapper) or Tensorflow.
  5. Again, you should find courses on Coursera and elsewhere to learn this.
  1. This and doing some projects available on Kaggle, on the internet should help you get practice.

You are a data scientist, Machine Learning, and Deep Learning Engineer at this point.*

*There is a lot more involved about how to handle big data and streaming data and other aspects. You can get this experience of handling large datasets on your job or by doing Kaggle projects.

Badge

Following this thread. Starting my course “Introduction to Data Science Specialization” today. 

Badge

Thank you all for your comments and suggestions, I will certainly draw from it and take on most of the suggestions you guys posted.  Thanks everyone!

Badge

Thanks everyone for their valuable insights about the field. It may provide with the route plan to take for a beginner. I'm also starting things from scratch.

Badge

I started data science course offered by johns hopkins

Badge +1

Hi,

I would suggest starting with Statistics fundamentals. You might have learnt some of them in high school - good to get them refreshed; some of the others it would be good to learn them afresh.

Try finding some beginner books on online bookstores or even Coursera might have some good courses.

Step 2 is doing a refresher on basic linear algebra. I think some videos on YouTube might help here.

Step 3 is picking up a programming language suited for DS. Most common are Python and R. Pick any and learn it and practice some basic data operations just to get the hang of it.

Step 4 is to do some DS courses. Coursera offers the DS specialization which looks good.

Step 5 is practice. Take some datasets offered for free on websites like Kaggle and see what you can do with it. Team up with others who have complementary skills and enjoy the experience.

 

Hope this helps!

 

Jayesh

 

Reply