11-15-2013 02:53 PM
Both products seem to have the capability to identify high risk patients for readmissions. My question is what are the general guidelines to choose one product over the other? What are things about my enterprise data model that would help determine which product would be best? Thanks.
11-15-2013 03:05 PM
If you're looking for statistical analysis, e.g neural networks,decisions trees, regression analysis I'd go for EMiner. I'd only for for VA if you need to create an interactive mechanism to explore the data though there are other tools within SAS that will do this.
11-15-2013 07:20 PM
You can't compare these 2 products in this way and it really depends on what business problem you want to solve.
It could well be that you're using EM to create models which are then used to create a data mart with indicator columns for risk or the like. You then load this data mart into VA for visualisation and investigation.
11-19-2013 02:11 PM
Thank you for being active in the VA community!
SAS Visual Analytics and SAS Enterprise Miner have some overlapping functionality, but are designed to work with the same data.
A suggestion is to consider the general skills and needs of your users, and the broader user community. Do you have other needs to support visualization, dashboarding or general reporting? If so, Visual Analytics might be a good fit. On the other hand, if your community includes advanced statisticians, SAS Enterprise Miner is a powerful tool.
Decision trees, regression analysis and neural networks are all very traditional statistical analysis functions. Visual Analytics can do linear regression analysis, decision trees, and basic network diagrams. SAS Enterprise Miner offers more depth around the functions you have described here. If a user is highly advanced, requires more functionality around the procedures described, and the user would like to publish the model to a data store then Enterprise Miner is likely the preferred choice. Enterprise Miner provides robust neural networks, allows you to tweak and compare multiple models and supports a broad palette of predictive analytics.
The products are designed to work together and by the end of the year, we expect to rollout new versions of both products. We are excited that a web-based mining product will be integrated with the SAS Visual Analytics Hub, enabling you to move easily between the functions. Your advanced analysts will have access to the tools they need (in a web-based environment), and your business analysts can share information, publish reports, and perform related analysis and explorations of the data.
06-05-2016 06:06 PM
i want to know the sas visual studion vs sas eminer for glm. how these two are different. and as i am replying to a old post so what changes have done in recent years ?
06-05-2016 06:38 PM
Since the previous post, SAS has released SAS Visual Statistics which may provide the GLM features you are after. For further fine tuning and sophisticated data mining, I'd look at SAS Enterprise Miner. It just depends on what features you are after. Here are a few links that may help you with your analysis:
SAS Visual Statistics product features - http://www.sas.com/en_ph/software/analytics/visual-statistics.html#m=features
SAS Visual Statistics documentation - http://support.sas.com/software/products/vs/#s1=2
SAS Enterprise Miner features - http://www.sas.com/en_ph/software/analytics/enterprise-miner.html#m=features
SAS Enterprise Miner product - http://support.sas.com/software/products/miner/
SAS Enterprise Minder documentation - http://support.sas.com/software/products/miner/#s1=3
If you have specific GLM requirements, I'd look at the documentation to see if they are met in the products.
11-20-2013 05:04 AM
At the moment we are using EG/DI to acquire and transform the data, EM/RPM to do all the modelling and pumping the results into VA to visualise and explore.
We are hanging out for an SAS R&D smack-down that results in full DI and EM capabilities being moved into VA ;-)