IBM SPSS Modeler Review

Saves us notable time in our go-live process


What is our primary use case?

We use it for data modeling like arithmetic modeling, bank modeling. We have different models such as loan models. We use three products, SAS, R, and SPSS Modeler to do predictive modeling. We are a big IBM shop.

I'm not sure how many machine-learning models we are putting into production. I'm new, I've been at the company for five months, but I would say this year there should be at least five or six models. We do a PoC on modeling and, based on what fits better, that's what we go with. So the bottom line is that a handful of models will go live but we'll be trying 10 to 15 models to do the predictions and see what best suits the company.

This is batch. We do monthly modeling, we do weekly modeling. It's not daily. We run weekly model reports too. We also change the parameters that we enter based on the industry, as things change.

We don't have cloud, it's all on-prem.

How has it helped my organization?

Our go-live process has changed compared to the previously programmatic code based process. It’s not just the time to go-live but it’s also the process itself; the improvement in terms of performance, and maintenance is also important. I would say it has saved us a lot of time, about 20 or 30% of our time. I don’t have the numbers in front of me but I think something along those lines.

What is most valuable?

We are big-time into data analytics. AI is another area which we want to start looking at. Digital banking is important. We are looking more into digital banking and we are trying to put some features in there. I think the trend is more on that area of data analytics, digital.

I can't comment on our use of SPSS Modeler for governance and security issues.

We use analytics with the visual modeling capability to leverage productivity improvements.

What needs improvement?

New features are always welcome, but I’m not the core person. A separate team can comment on this, but not me.

What do I think about the stability of the solution?

There are issues, we try to mitigate them. There are always issues. We’re trying to be stable but there are a few areas...

What do I think about the scalability of the solution?

It’s definitely scalable, it’s all on the same platform, it’s well integrated. I think the integration is important in terms of scalablility because essentially, having the entire suite helps a lot to scale it, market it. Even in terms of processing, it’s easier.

How is customer service and technical support?

I personally have not had experience with IBM technical support, but the group has worked with them. I haven't heard anything from them, so I think it's okay.

Which solutions did we use previously?

We already had SAS, we had R. It’s all legacy and it’s all homegrown. But we had an IBM shop also.

What other advice do I have?

I would say, look through every product in the market, like we do, and try to pick what works best.

Disclosure: I am a real user, and this review is based on my own experience and opinions.
Add a Comment
Guest
Sign Up with Email