Big Data Analytics, Data Mining, Machine Learning

Becoming a Data Scientist – Curriculum via Metromap

Data Science, Machine Learning, Big Data Analytics, Cognitive Computing …. well all of us have been avalanched with articles, skills demand info graph’s and point of views on these topics (yawn!). One thing is for sure; you cannot become a data scientist overnight. Its a journey, for sure a challenging one. But how do you go about becoming one? Where to start? When do you start seeing light at the end of the tunnel? What is the learning roadmap? What tools and techniques do I need to know? How will you know when you have achieved your goal?

Given how critical visualization is for data science, ironically I was not able to find (except for a few), pragmatic and yet visual representation of what it takes to become a data scientist. So here is my modest attempt at creating a curriculum, a learning plan that one can use in this becoming a data scientist journey. I took inspiration from the metro maps and used it to depict the learning path. I organized the overall plan progressively into the following areas / domains,

  1. Fundamentals
  2. Statistics
  3. Programming
  4. Machine Learning
  5. Text Mining / Natural Language Processing
  6. Data Visualization
  7. Big Data
  8. Data Ingestion
  9. Data Munging
  10. Toolbox

Each area  / domain is represented as a “metro line”, with the stations depicting the topics you must learn / master / understand in a progressive fashion. The idea is you pick a line, catch a train and go thru all the stations (topics) till you reach the final destination (or) switch to the next line. I have progressively marked each station (line) 1 thru 10 to indicate the order in which you travel. You can use this as an individual learning plan to identify the areas you most want to develop and the acquire skills. By no means this is the end; but a solid start. Feel free to leave your comments and constructive feedback.

PS: I did not want to impose the use of any commercial tools in this plan. I have based this plan on tools/libraries available as open source for the most part. If you have access to a commercial software such as IBM SPSS or SAS Enterprise Miner, by all means go for it. The plan still holds good.

PS: I originally wanted to create an interactive visualization using D3.js or InfoVis. But wanted to get this out quickly. Maybe I will do an interactive map in the next iteration.

105 thoughts on “Becoming a Data Scientist – Curriculum via Metromap

  1. Swami this is great visual treat!
    In your next iteration you can consider super-imposing Big Data and Visualisation as layers on top of the basic layers !

    Another iteration would be to present the info as relevant to the people in the org – starting with the developer / data miner all the way up to the CIO/CEO.

    All the best.

  2. Nice chart!
    Maybe you can provide a printer-friendly version, inverting black and white colors. This way, people will be able to print it, and annotate the “stops” where they have already been.

  3. Swami-
    I am beginning my path to becoming a data scientist – have my undergrad in Mathematics and trying to determine where to go next and how to do it.
    In addition to the MBA I am going to begin pursuing, this map is very useful. I have been looking for this type of thing for a few days now, and yours tops everything I’ve seen.

  4. Excellent work and highly remarkable. This helps everybody who wanted to pursue their carrier in Data Scientist.

  5. Swami,
    You have done a great work that is defiantly going to help people who are willing to explore the field of Data Science.
    I have a query, can you explain, what does the %age values represent in the road map (Shown inside yellow stars)?

    Thanks in advance,

  6. Thanks for sharing this journey map. A few things come to mind when exploring the map:
    1/ Not every station should be equal – you can’t be good at every skill. Data scientist is team work with different roles. I suggest a “High Lights Line” across the Main Stations of each line. These stations represent the essentials which every data science team player needs to know.
    2/ The line travelled depends on the role of the traveller in the data science team – Not every line needs to be travelled by every team member. Each member travels the line which is relevant for his role (after he/she has first travelled the High Lights Line).
    3/ There is (currently) no line for soft skills – very important to communicate/present your work and understand where other people are coming from when talking about data. You can be the best data sciencist out there, but if you can’t convey your message, your work will not be resognized. Social skills are pretty important in team work.


  7. Swami,

    I believe that I can say that in the name of your target audience: this is a great help, we love you, but two descriptive sentences on each stop could exponentially increase the effective usefulness of this map.

    The reason is that we do not know yet the domain, so while titles are fine mnemonics for experts, they are just not enough to get started for a beginner.

    Thank you again,

  8. Thanks all for your kind words, suggestions and comments. I will incorporate these in my next version. Happy that I was able to make a small dent and difference in this (Data Science) hot and yet ambiguous area.

    Also I’m working on zooming into each of the domains & corresponding stops to provide a 1-2 sentence overview of what a particular topic means. Kind of like a visual cheatsheet.

  9. Wow, this is really great. This is going to start appearing on cubicle walls very quickly!! It’s great for data scientists that want to round off their skills.

  10. Hello Swami

    Thanks a lot for visualization the map of Data Scientist.

    Could you please provide us some referance for all this stations (topics).

    Himanshu Jha

  11. This really is helpful! I find that I started in the middle of the path (stations 4-6) and I am struggling to understand what I need from stations 1-3 and what to do to get through stations 7-10. Can you post what sources you consulted to make this? I think that would help me understand it better.

  12. Loved the way you have captured it along with percentage completion. Would it be ok with you if I used this pic in conveying a message? Credits will be duly given, rest assured.

  13. Hi Swami,

    Great Article… This is the only article I found in the internet which clearly describes the information about what a Data Scientist should know !

    If possible can you suggest some books ?

    Thanks in Advance.
    Mahesh Chandrasekaran

    1. – If you want to get applied skills, and if you have picked the R route, I would recommend “Data Mining with R: Learning with Case Studies (ISBN-10: 1439810184)” and go through it end to end.
      – If you have access to IBM SPSS Modeler then use the Applications Guide,
      – There are tons of data sets that are publicly available – UCI, Amazon Review Data, etc. Use them and DIY.
      – Also recommend “Machine Learning for Hackers” (ISBN: 1449303714)

  14. Don’t know why it has suddenly become a dirty word but good old fashioned SQL is still going to be required for quite a while. Believe it or not most Data Scientists are going to run into it sooner rather than later.

  15. The Metro-map is superb. Quite exhaustive. But it took me some time to digest the cumulative percentages. This is my next favourite Data Science visual depiction after the Data-science Venn diagram of Drew Conway.(in order of chronology only)

    Thanks for putting it up.

  16. Hi Swamy ,

    This is an eye opener . Recently I graduated to become a data scientist in my organisation and i have realized that there are multiple things that I have missed in my journey and they do seem inevitable looking at the links here.

    This is really awesome !

    Anirudh Kala

  17. Hi Swami,

    I’ve heard about this map from a friend of mine. That is really a wonderful work. Thanks.

    Is it okey if i use the map by addressing through this page on the capture in my blog?

  18. Brilliant Swami. This will help me to be on the right track and change tracks appropriately. I will have this on my wall.

  19. Mr.Swami, really good work!
    This is abosolutely wonderful illustration to map high level learning abstractions into low level grains.

  20. Would I really just become a data scientist or get some credits with universities i would enroll in so as to officially graduate, if I just worked on material/resources required?

    Thanks again Shwami

  21. Hi Swami, really impressive visual. I am studying to get into data science at the moment so this is incredibly helpful. any chance of emailing me a high quality version, i would like to print it onto a large poster and put it up on the wall in my room. Thanks, Hom

  22. Outstanding!

    I came across your page while searching for a framework to train data scientists. This is the most comprehensive and cohesive one I have come across so far. The potential for extending this is enormous as pointed out in a bunch of replies. This can easily become a framework using which users can add their own weights, skills tracks, highlights, checklists, etc!

    Hoping that you get time to complete the D3 version!

    Thanks for sharing. I am hoping that the permission you granted to “Eroteme” applies to all of us.

  23. I cannot express in words my gratitude for putting this together. As a current programmer looking to venture in data science – this has provided an excellent roadmap for continued learning / working towards my goal of being a full time data scientist.

    Thanks again,

  24. Swami,this is nice job.
    In programming, why did you find it necessary to have both Python and R?can’t one of these (or any other e.g. Octave, C++) be sufficient?
    Nice time.

  25. Nice pic!

    Only one thing: IMHO Support Vector Machines is missplaced. It should be in the Machine Learning path.

  26. I think this is a great technical roadmap, but where is the business domain tube line or would we continue the metaphor and say each city in which this tube map is used, is a business domain/context: London city = Customer, Munich = Operations/Supply Chain, New York City = Risk, Tokyo = Design. The business context is one of the most key areas for applying data science skills and adds that extra dimension that moves an academic data scientist from the one dimension mathematical/scientific into the application to business context, thus rounding out the profile.

  27. Hi

    Very useful map. Thanks.
    By the way, this was published two years back.
    Any updates you would suggest in the map??

  28. Hi, I would like to thank you for publishing such a good road map. Can you however, guide me on how to get started because I’m a fresher with absolutely no experience. Can you suggest the requisite courses to be done and skills to be learnt to successfully become a data scientist?

    Thank you again 🙂

  29. Hi Swami,

    Thank you for the excellent work!

    A question please, this visualisation shows how to become a Data Scientist getting on the different trains each time and learn the required skills. But does this visual also shows that these are the stages a project should follow? So, would each one of the Data Science roles reflect in one of these lines?

    Fundamentals :
    Statistics : Statistician
    Programming : Data Engineer / Data Architecture?
    Machine Learning
    Text Mining / Natural Language Processing
    Data Visualization
    Big Data
    Data Ingestion
    Data Munging

    Please if anyone knows can you fill in what job does what part of the chart?

    Will be greatly appreciated!


  30. Hi Swami,

    Great job on this! Would it be possible to get a hi res copy of this graphic? Would like to print and laminate for my team’s war room to inspire them as they become more skilled in data science.

    Please PM me if this is possible and I have your permission to print and laminate.


  31. Hi guys,

    I just created a Github to work with this beautiful roadmap.

    We could start posting tutoriels, short pieces of code .. Everything to make it more clear for new people in data science.

    I’ll start to add some, but I can’t do this all by myself 😉

  32. Hi Swami,

    I am conducting research about using diagrams in searching and browsing. Can I use your diagram about data science? I may include it in a paper or in a demo website with a complete citation (I will include your name and url).

    Thank you.
    Hisham Benotman.

  33. Hi, Swami
    Great job! Is the interactive version of this Data science subway map already available ? Can I use your map on my website including your name and url ?
    Thank you very much.

  34. your post is too good and those who are looking for making career in data science, for them this post is mind blowing,if they start learning from here really they are going to change future and make good out of it. Thanks for sharing this on Data science ,keep doing it and really it is very helpful.

Leave a Reply

Your email address will not be published. Required fields are marked *