Resolved: Your inforgraphic is wrong
Please share the sources or your information.
it seems that you are wrong about machine learning, You have classified regression, classification and clustering as not ML, but they are the basis of most ML.
NN and SVM both work with these concepts and as I am not very familiar with other terms they may be as well.
so, the best classification would be to include regression, classification and clustering as ML and the algorithms as Deep Learning. you may reference MS Learn and the IBM cloud education resources to correct this course.
i am very disappointed in this course and believe that a better effort must be put in to ensure current and accurate information. People who are learning for the first time will learn the woncepts wrong.
Hi Balwinder!
Thanks for reaching out.
Indeed, all coming from academic background, we are advocates of preserving the evolution of the meaning behind the terms used and the disciplines that have appeared over time.
Machine learning uses regressions and classification, as you say. But these are statistical phenomena that have been around way earlier than the term "machine learning" came around.
Same is valid for the term "algorithm".
We do understand the confusion and that's what motivated us to create this Introduction to Data and Data Science course. We would like you to be aware of the terms and buzzwords as they appeared historically to better understand what each discipline is about, as well as why and how we can take advantage of each one of them.
If you have other questions you'd like to ask, please don't hesitate to reply to this message or open up a new thread in the Q&A.
Thank you!
Hope this helps.
Kind regards,
Martin
I agree with Martin the instructor, and no, I don't work for 365, this is actually my first time taking the course and I'm only 5 lessons in. But I was actually blown away by this infographic and would agree 100% that it's breakdown is spot-on. I think part of the confusion for some students (particularly those who are younger, no offense..) as that many now (hard to believe) are actually growing up in a time where machine learning always existed for them. Now of course, true machine learning has been around for decades, so I'm not trying to get into the weeds of that argument about what true 'machine learning' is, lest I get into beef with some of the old heads. But you know what I mean, machine learning in the modern sense that most people understand it today with all of the modern algorithms, i.e. ML that has come along with the advent of Big Data, in the last 10-15 years. Having said that, I went to college in the mid 90's and remember all of the 'traditional' methods (regression, etc.) taught to us in STATs 101 which we had to compute all by hand and with graph paper. (yes there is still such a thing as graph paper ;) ) And all of those methods existed even before relational databases had even matured or Windows 95 even existed. They may be a basis for a lot of ML, but they are just basic stats, they are not ML. (Or maybe more accurately put, *most* ML is just applied/advanced stats a computer is managing on its own.)
Hi BB!
Thanks for reaching out and joining this thread by sharing such a valuable story and information! Thank you very much for the kind words, also.
... Of course, I hope you'll find the entire course and Program beneficial to you. If you need any assistace, please do not hesitate to post a question below or in a new thread, we'll be glad to help. Thank you.
Kind regards,
Martin
Sure thing. Looking forward to it, I actually just purchased an additional year. This is filling in a lot of blanks for me. I realize everyone learns differently so by no means is my 'old school' method better or worse than anyone elses, but I find it most helpful while doing these courses (particularly the stats portions which constitute the base of most of the rest of it) to do all of the calculations by hand on paper, pen and simple calculator. Almost every single formula that we are reliant on Python to calculate for us through libraries can be calculated by hand. The advantage of this approach is that once you know how the formulas work, you know it. All the computer is doing for you is crunching 100s and 1000s of these calculations for you much faster.