@vivekps wrote:
Hi,
I am just going to give a brief summary of how exactly you should choose a course to pursue data science and business analytics.
Introduction:- I completed a 1 year program with one of the acclaimed ranked program in this country. I would like to emphasize on the words acclaimed and ranked which one should always take with a pinch of salt as there is more to a university than these ranks.
Why did i chose what i chose:- I didnt not have a chance to read a post like this.
So to the question what should i look for in a course before choosing it:-
1) Faculty - Get complete list of faculty for the full course. Check their credentials. Even if the schools claims to have used industry leaders, validate if these leaders are actually leaders or just someone who is working in Data science and is as new as you are if not for that 1 year at work. Go view their Linked in.
2) Statistics faculty - Forget about these jargons of machine learning and what not, for the core statistics and advance statistics is what you should be looking at. Reason, most of the concepts on technique is freely available for copy paste and many industry leading faculties i have had a chance to work with in the school are basically at same level as someone doing 2-3 kaggle projects copy pasting those codes.
Point 1 and 2 i am emphasizing on faculty because most hands on industry self acclaimed experts are absolutely poor to an extent of useless in teaching. As much as machine learning is a skill set and some have the tendency to be masters in it, teaching is also a specific skill that comes with training, practice and experience. So Any school that claims to have industry, live project etc with no proper faculty will be as good as reading forums, articles and doing your own coursera course while not wasting money in any of these schools.
3) Placement from a known bigger named school will always be better than lesser known higher percentage showing schools in terms of placements. In other words, there is a reason why IIM have their brand value. Choose brand. Do not choose a school simply by rank. The branded ones have multiple advantages that these non branded school cannot offer.
a) Branded schools bring in the best of students. PERIOD. Meaning you get better alumni connect and people who can actually get you a project or change of project in long run.
b) Branded schools attract better companies and can prepare you better for those jobs.
c) Non Branded schools themseleves are trying to find a feet on ground and even faculty quality is sub standard in most of them.4) These courses will not prepare you straightaway to become data scientists. Data scientists are probably PHD's who know the right way to interpret the data than those going for a 1 year crash course on it. What these courses will help you is to get you hands on in technology and a very brief over view of how this can help industries.
5) Dont fall for ranking. Reach out to alumnis and ask for realistic feedback.
6) If there is a new branch of school that is opening, do not take it. They dont have infrastructure nor faculty structure. You will waste your money. I know about it by experience.
Part time vs Full time
It doesnt matter. My advice save your money by having no break in compensation by doing a weekend or part time program. My brother did in Great Lakes and the faculty did make a difference and the topics and depth of topics covered was as good as in the full time course i did.
My advice - Take part time, apply in your line of work. At the end the degree from these schools DO NOT Carry major value. Instead what projects you worked on, techniques you worked on and the ability to talk in those terms matters when it comes to actual job. The companies wont even know the name of the schools you went to if its non branded.
Finally,
Stay relaxed. Reach out to people who studied the program. Ask for advice from different sources before jumping. If there is a branded program and its costlier, it will still have better ROI than non branded schools.Hope this helps prospective students.
Posts: 1
Participants: 1