Mainframe, MVS and zOS Discussion

Hi,

Can Data science be praticed in mainframes? I am currently working in to mainframes and would like to move to data science, what is the best wayh to learn it and be in mainframes too?

What is "data" science (or data "science")?

To some degree, data science is independent of the platform, so it can be done on mainframes or Unix servers or Windows servers or whatever. Unless you are interested in getting a doctorate at a university, I suspect you probably don't want to move to data science but into some aspect of data science (data analytics, statistics, or whatever).

nicc wrote: What is "data" science (or data "science")?

Google it, wikipedia has the information on it: Data science is an interdisciplinary field about processes and systems to extract knowledge or insights from data in various forms, either structured or unstructured, which is a continuation of some of the data analysis fields such as statistics, machine learning, data mining, and predictive analytics, similar to Knowledge Discovery in Databases (KDD).

Robert Sample wrote: To some degree, data science is independent of the platform, so it can be done on mainframes or Unix servers or Windows servers or whatever. Unless you are interested in getting a doctorate at a university, I suspect you probably don't want to move to data science but into some aspect of data science (data analytics, statistics, or whatever).

Thanks. But what is the correct way go for it? Should I invest my time leaning SAS? Will only SAS be enough? Should I also learn Python or R? What else? Do I need to learn some tool for this?

But what is the correct way go for it? Should I invest my time leaning SAS? Will only SAS be enough? Should I also learn Python or R? What else? Do I need to learn some tool for this?

The CORRECT way would be to figure out you want to do; this will drive what tools you learn because statistical analysis uses different tools than data mining and than other aspects of data science. SAS and R are statistical packages and they have their place in data science -- I'm not sure why you compare them to Python. But other tools can be handy -- Google data mining tools and you'll get back about 9,920,000 hits for example. If you decide to learn SAS and you wind up working for a company that doesn't have it licensed, why did you learn it?

Thanks for all the answers. I'm trying to get more information form your experience, though I might not be very correct in outting up all my questions well.

That's right that if I learn SAS and company is not using it then it will be of no use but then it applies to COBOL too. I was in a project which made use of only PL/I and I had knowledge of COBOL so I learned somewhat L/I too.

That's right that if I learn SAS and company is not using it then it will be of no use but then it applies to COBOL too. I was in a project which made use of only PL/I and I had knowledge of COBOL so I learned somewhat L/I too.

The point we are making is that if you focus on the tools, you are limiting yourself more than if you focus on the underlying techniques and algorithms.

Thanks Robert.

But as I'm working in mainframes so I was looking forward to start with some tools which can be used in mainframes and that will help me going forward.....

But as I'm working in mainframes so I was looking forward to start with some tools which can be used in mainframes and that will help me going forward

This is a COMPLETELY different question than asking about data science. Start by finding out which tools are installed AT YOUR SITE and find out how to get the training to be able to use them. Or you may want to do some Google searches (hint: mainframe tools is a good way to start looking) to narrow your search down. However, whether or not any particular tool (or method or technique or anything) will help you going forward is a question that only you can answer -- and that should be driven by your interests, not whatever the market place is currently doing.