That's a great job done with this whole series of introductory videos about SAS EM!
@SASSoftware
9 жыл бұрын
+Radoslav Dzhidzhov Thanks so much for your feedback! We're glad to hear this!
@adithyavarambally6494
4 жыл бұрын
it couldnt be explained any better ..thanks SAS
@SASSoftware
4 жыл бұрын
Awesome! Glad you found it helpful!
@raquelgve
9 жыл бұрын
so useful, thanks!
@harikas.8806
8 жыл бұрын
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?
@SASSoftware
8 жыл бұрын
Hi +Harika S. Thanks for reaching out! We're looking into this one for you!
@SASSoftware
8 жыл бұрын
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.
@enssea.hamimed
2 жыл бұрын
How about Mixed models?
@dr.abhijeetsafai7333
5 жыл бұрын
I am not able to read a file from physical drive by SAS Enterprise Minor because the library is created at server as the software is running at server. May I request you to kindly share how I can read a file by libname or fiename statement at physical location in this case? Thanks in advance. - Dr. Abhijeet Safai
@SASSoftware
5 жыл бұрын
Let us know if this helps: 2.sas.com/6057EzKdv If you're not able to resolve your issue, we'd recommend posting to the SAS Communities where experts are available to assist with usage/programming questions.
@rohitsinha2439
7 жыл бұрын
Hi, this video is very informative. But I wanted to know why the role of TargetD is set to 'rejected'.(10:17)
@SASSoftware
7 жыл бұрын
Rohit, thanks for your inquiry! We are looking into this for you.
@SASSoftware
7 жыл бұрын
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!
@jeff6830
5 жыл бұрын
Can you add the link to the sample data in the description pane?
@SASSoftware
5 жыл бұрын
Jeffrey, thank you for reaching out to us! We are looking in to this for you. Note you can also find additional resources here 2.sas.com/6052ENR8Q
@SASSoftware
5 жыл бұрын
Jeffrey, We recommend checking the documentation link for Enterprise Miner, here: 2.sas.com/6055ENRZ3 and then look for the second bullet on the top of the page, entitled: "Example Data for Getting Started with SAS Enterprise Miner 15.1" at the end of that bullet is a link to download the ZIP file for the Enterprise Miner sample data.
@oloratomotitswane1123
6 жыл бұрын
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??
@SASSoftware
6 жыл бұрын
Olorato, thank you for your inquiry! Did you try downloading the zip file here?2.sas.com/6055D4raU
@oloratomotitswane1123
6 жыл бұрын
Yes i did
@oloratomotitswane1123
6 жыл бұрын
Thank you so much...I just downloaded the zip file using that URL and it worked..Thanks alot
@SASSoftware
6 жыл бұрын
Olorato, glad it worked for you!
@rmadzikanda
7 жыл бұрын
can we combine two or more separate datasets
@SASSoftware
7 жыл бұрын
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." 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? 2.sas.com/60548Xgin We hope that helps!
@chelovekx121
8 жыл бұрын
It's not clear how correctly write full network path to the client from video
@SASSoftware
8 жыл бұрын
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!
@filmzone46
4 жыл бұрын
Hum ... i did not find the sample data :(
@SASSoftware
4 жыл бұрын
Hi there! We recommend checking the documentation link for Enterprise Miner, here: 2.sas.com/6055ENRZ3 and then look for the second bullet on the top of the page, entitled: "Example Data for Getting Started with SAS Enterprise Miner 15.1" at the end of that bullet is a link to download the ZIP file for the Enterprise Miner sample data.
@filmzone46
4 жыл бұрын
SAS Software amazing !!!!!
@SASSoftware
4 жыл бұрын
Thanks for sharing, we're glad you enjoyed the content!
Пікірлер: 33