Watch this Ask the Expert session to learn data-driven, self-service data preparation techniques like transforming, blending, shaping and creating computation variables (feature engineering).
Watch the webinar
Join John Stultz as he discusses how to use AI and machine learning to scan data and make intelligent transformation suggestions, create prebuilt transformations and data cleansing functions, share automatically generated code with IT and data preparation jobs as centrally managed assets. Watch this webinar to learn:
How to move from "80% preparation and 20% analysis" to "20% preparation and 80% analysis."
To create explanatory variables (feature engineering) that can improve model (machine learning) performance.
Ways to improve confidence in data, increasing accessibility and trust in modeling.
The questions from the Q&A segment held at the end of the webinar are listed below and the slides from the webinar are attached.
What type of examples of computational variables are common with fraud detection techniques?
Depending on the data and type of behavior you are trying to detect, the key is having someone with domain expertise to create the best computational variables. You want to think about how you can identify suspicious behavior in ways that flag or quantify behaviors. For example, peer group analysis is a great way to do fraud anomaly detection using ratios. Binary flag variables, 0 or 1 is another way. Flagging if something is happening in a period of time is also a good detection technique. Basically quantifying behaviors that are inconsistent.
What do I need to license in order to be able to prepare my data using SAS Data Preparation?
SAS Data Preparation is bundled into a lot of the different product suites we have at SAS. The best way to figure out is to reach out to your Account Executive or Customer Success Manager to look at your current license or discuss what suite can meet your needs.
I used to run a PC SAS shop, and now I use RGUI. I'm having trouble seeing how what you are demonstrating is less clumsy than either of those. Perhaps I am missing something?
There are ways to leverage R as an open source capability to drive the modeling techniques. Feel free to comment below if we haven’t answered your question.
The transform process seems similar to the SAS EG Import Data steps when the user defines field attributes. Is that true?
Yes, in terms of defining field attributes it is a similar process.
I think I forgot what he mentioned for SAS Studio he is showing. Is this for either SAS9 or Viya?
It’s for both. You can have a SAS Studio license and run it in a SAS 9 installation or a Viya installation.
Recommended Resources
Five Data Management and Analytics Best Practices for Becoming Data-Driven
Data Preparation Challenges Facing Every Enterprise
Be Prepared: An Introduction to SAS® Data Preparation
5 Data Management for Analytics Best Practices
Save 50% on SAS Certification Exams
Want more tips? Be sure to subscribe to the Ask the Expert board to receive follow up Q&A, slides and recordings from other SAS Ask the Expert webinars.
... View more
