HomeScience & TechnologyRelated VideosMore From: SAS Software

Getting Started with SAS Enterprise Miner: Setting Up an Enterprise Miner Project

128 ratings | 76957 views
http://support.sas.com/software/products/miner/index.html Chip Robie of SAS presents the first in a series of six "Getting Started with SAS Enterprise Miner 13.2" videos. This first video demonstrates how to set up an Enterprise Miner project. For more information regarding SAS Enterprise Miner, please visit http://support.sas.com/software/products/miner/index.html SAS ENTERPRISE MINER SAS Enterprise Miner streamlines the data mining process so you can create accurate predictive and descriptive analytical models using vast amounts of data. Our customers use this software to detect fraud, minimize risk, anticipate resource demands, reduce asset downtime, increase response rates for marketing campaigns and curb customer attrition. LEARN MORE ABOUT SAS ENTERPRISE MINER http://www.sas.com/en_us/software/analytics/enterprise-miner.html SUBSCRIBE TO THE SAS SOFTWARE YOUTUBE CHANNEL http://www.youtube.com/subscription_center?add_user=sassoftware ABOUT SAS SAS is the leader in business analytics software and services, and the largest independent vendor in the business intelligence market. Through innovative solutions, SAS helps customers at more than 75,000 sites improve performance and deliver value by making better decisions faster. Since 1976 SAS has been giving customers around the world The Power to Know.® VISIT SAS http://www.sas.com CONNECT WITH SAS SAS ► http://www.sas.com SAS Customer Support ► http://support.sas.com SAS Communities ► http://communities.sas.com Facebook ► https://www.facebook.com/SASsoftware Twitter ► https://www.twitter.com/SASsoftware LinkedIn ► http://www.linkedin.com/company/sas Google+ ► https://plus.google.com/+sassoftware Blogs ► http://blogs.sas.com RSS ►http://www.sas.com/rss
Get embed code!
Text Comments (21)
Olorato Motitswane (13 days ago)
I am trying to create an enterprise miner data source using sas 7.1, under selecting a sas table the table donor.donor_raw_data does not exist or could not be found..What do i do??
SAS Software (12 days ago)
Olorato, glad it worked for you!
Olorato Motitswane (12 days ago)
Thank you so much...I just downloaded the zip file using that URL and it worked..Thanks alot
Olorato Motitswane (12 days ago)
Yes i did
SAS Software (13 days ago)
Olorato, thank you for your inquiry! Did you try downloading the zip file here?http://2.sas.com/6055D4raU
Samuel Jackvony (10 months ago)
Thank God for Chip!
SAS Software (10 months ago)
Indeed! We think he's pretty awesome, too!
rohit sinha (1 year ago)
Hi, this video is very informative. But I wanted to know why the role of TargetD is set to 'rejected'.(10:17)
SAS Software (1 year ago)
Hello again, Rohit! The simplest answer to your question is found in the original statement of the data mining problem: TARGET_B is the binary event we want to model. TARGET_D contains different data, so we do not need it for this scenario. TARGET_B contains the binary data indicating whether an individual donated (1) or not (0). We want *all* donations, not just large donations, so we’re interested in TARGET_B=1, donation happened. In other words: All donations > 0 are what we want. The variable named TARGET_D sounds interesting, but the exercise says to discard it. This is because the data in TARGET_D is not the binary event we want to model. TARGET_D contains metadata about donation size, which is a decision that occurs in time *after* the event we are modeling (binary TARGET_B=1) happens. Again, this means we don’t need TARGET_D because it isn’t our objective for this problem. A slightly different way to explain this: Inspecting the raw data in DONOR.DONOR_RAW_DATA shows us that TARGET_D is a numeric variable that represents the donation amount in dollars, if a donation was made. However, a donation amount is decided only after a donation event, TARGET_B=1, has already occurred. Because the state of TARGET_D can only have values after event TARGET_B=1 has already happened, TARGET_D data are not useful for predicting TARGET_B events. TARGET_D is not the event we want to model. TARGET_D also does not contain values that are useful input predictors for the TARGET_B event that we do want to model. Therefore, the variable TARGET_D was rejected from the analysis since it has no useful input or target role. TARGET_D *could* be used analytically in a different scenario than the one we have here. For example, the data in TARGET_D was used to calculate the average donation of $15 that was provided to users as a “given” during the Decision Weights step of the Data Source Wizard. With a different data mining assignment, the data in TARGET_D could be useful. For a list of all potential donors who were likely to not only donate, but also donate sums of $25 or larger, a data miner could use analytic tools such as SAS Enterprise Miner’s Two Stage Modeling node to generate lists of candidates most likely to 1) choose to donate *AND* 2) donate a sum larger than $25. This scenario, and Two-Stage modeling is beyond the scope of the Getting Started with SAS Enterprise Miner introductory data mining analysis. We hope that helps, and thanks for watching our videos!
SAS Software (1 year ago)
Rohit, thanks for your inquiry! We are looking into this for you.
Ronald Madzikanda (1 year ago)
can we combine two or more separate datasets
SAS Software (1 year ago)
Hello Ronald! Yes, data sets can be combined in SAS Enterprise Miner. According to our documentation, you can "use the Merge node to merge observations from two or more data sets into a single observation in a new data set." http://2.sas.com/60518XgiT Please see this SAS Support Communities post for more information In Enterprise Miner, can I merge two datasets by a variable in the role of ID? http://2.sas.com/60548Xgin We hope that helps!
Евгения Хуснутдинова (2 years ago)
It's not clear how correctly write full network path to the client from video
SAS Software (2 years ago)
Hi there! Your SAS administrator (or IT Help Desk) will be able to provide the network path from your site's SAS server to SAS client. If you're not sure who your on-site admin is, please email us your site number at SAS_Cares@sas.com and we can get you their info!
Harika S. (2 years ago)
Thanks for the tutorial. I have a question: What will happen if we exchange Decision 1 and 2 (change their order and the corresponding values)? How the model building process or scoring will be altered?
SAS Software (2 years ago)
Hi +Harika S., if you change the values that correspond to the benefit of hitting the target (1), or loss of missing the target (0) for one decision or another, then that will affect the analysis.  However, if you’re asking what happens if the columns for the variables are switched, and the weights for Decision 1 and 2 are left the same for a hit or a miss, then that shouldn’t change anything. The results might display differently, but they’ll be numerically identical.
SAS Software (2 years ago)
Hi +Harika S. Thanks for reaching out! We're looking into this one for you!
Radoslav Dzhidzhov (2 years ago)
That's a great job done with this whole series of introductory videos about SAS EM!
SAS Software (2 years ago)
+Radoslav Dzhidzhov Thanks so much for your feedback! We're glad to hear this!
SAS Software (3 years ago)
We appreciate your feedback, Raquel!
Raquel Gómez (3 years ago)
so useful, thanks!