About M_Maldonado

M_Maldonado · ‎11-23-2015

Hi Nehac, If you don't have the contact information for your SAS representative handy, you can create a contact form for a quote here: http://www.sas.com/en_us/software/how-to-buy/request-price-quote.html I don't remember top of my head what version first includes link analysis node, but 14.1 has a lot of other stuff you will be interested including HPForest and all the HP nodes. Here one of the best papers on link analysis: http://support.sas.com/rnd/app/data-mining/enterprise-miner/papers/2014/linkAnalysis2014.pdf Good luck! -Miguel

M_Maldonado · ‎11-23-2015

Your code seems OK, but for some reason EM thinks that you have not run your neural network yet: 8749 proc freq data=&EM_IMPORT_DATA; ERROR: File EMWS3.NEURAL_TRAIN.DATA does not exist. If you have this flow: Data(node)-->Neural network(node)--> sas code(node) click on your SAS Code node, go to Imported Data and browse or explore the training data. Confirm that data set exist, and your code should run. good luck! -M

M_Maldonado · ‎11-22-2015

Hi Matin, So what exactly are you trying to do? You mentioned that you have a SAS code node at the end of your flow. What are you trying to do with that SAS code? Hopefully there is a node to do what you are trying to do (this is the case 95% of the time). Walk us through what you are trying to achieve and we'll help you debug the code you wrote on the SAS code node. Thanks, Miguel

M_Maldonado · ‎11-20-2015

Hi Judy, The English Rules are stored as a text file on a subdirectory with the node ID of your tree. For example, for my decision tree number 3 on a diagram, the English Rules are in: D:\EM\EM_Projects\EM14.1\miguel\demo\Workspaces\EMWS1\Tree3\ENGLISHRULES.txt Once you know the physical path of your enterprise miner workspace (EMWS), the rest is pretty straightforward. I hope you get to share with everybody this automated program that you are doing! Good luck! -Miguel

M_Maldonado · ‎11-20-2015

The assessment in Enterprise Miner just lets you see the fit statistics for the test partition on the Model Comparison node. I think this is to encourage you to tweak your model to improve the statistics of your validation set, and hold the test partition for unbiased comparisons. Short answer, add a model comparison node. Good luck!

M_Maldonado · ‎11-20-2015

Hi, One of the main advantages of decision tree algorithms is that they handle missing values. If you still want to exclude missing values, you are doing it the right way. The reason you see a label for "or missing" is that it is part of the algorithm to assign missings (if there were any) to a specific branch. In other words, your decision tree is just telling you what it would do if you use this model to score a new data set that has that variable missing. A quick way to confirm that your filter did exclude missings: Select your decision tree node, click on the ellipsis for Imported Data, and browse or explore this partition to confirm that the observations with missings were excluded. Personally I would not exclude observations with missing values, but you can certainly do that if that is your preference. Good luck! -Miguel

M_Maldonado · ‎11-19-2015

Hi Benjamin, Check out the example in the reference help, and maybe that gives you a better idea of how this node works. Press F1 when you are on Enterpise Miner. Once in the Reference Help look for the secition "Merge Node Example". good luck! -Miguel

M_Maldonado · ‎11-19-2015

Hi Jakes, When you say you see the extra variables in the score code, are you sure that you are looking at the score code from HPForest? Random Forests in Enterprise Miner is one of the few procedures that do not produce SAS Score code. The score code would be quite large, so instead your HPForest node produces a file that another procedure (proc hp4score) uses to score. If you go to results to see the SAS code, you will see something like the below proc hp4score data=&hpfst_score_input; id &hpfst_id_vars; %if %symexist(EM_USER_OUTMDLFILE)=0 %then %do; score file="D:\EM\EM_Projects\EM14.1\miguel\demo\Workspaces\EMWS1\HPDMForest2\OUTMDLFILE.bin" out=&hpfst_score_output; %end; %else %do; score file="&EM_USER_OUTMDLFILE" out=&hpfst_score_output; %end; PERFORMANCE DETAILS; run; Kinda puzzled here. Do you have a screenshot that suggests that HPForest is not honoring your metadata selections? Thanks, Miguel

M_Maldonado · ‎11-17-2015

Hi Albe, Since your diagram is simple and your "Best" gradient boosting node is running OK, let's try to use the score code manually. First, identify the ID of your gradient boosting node (click on it, and on top of their properties find the Node ID e.g. Boost3). Do the same for your workspace, click anywhere blank on your diagram and find the ID (e.g. EMWS1). Next, locate the physical path of your workspace. One way to do it: click on the explorer icon, select Show Project Data, and confirm the physical path. Now go to base SAS and use code similar to the below. I mocked up my test data, but you can start on the section Score manually. /*Create a test set. To mock it up, I dropped the target variable*/ data test; set sampsio.hmeq; drop bad; run; /*Score manually*/ %let path=D:\EM\EM_Projects\EM14.1\mim\demo\Workspaces\EMWS1; %let nodeid=Boost3; filename emflow "&path.\&nodeid.\EMPUBLISHSCORE.sas"; data test; set test; %inc emflow; run; Hopefully scoring manually gets you somewhere? Also, if you have time, send me the log of the Score node that is failing. When you are on the results, go to View-SAS Results-> log. Do CTRL+F to try to find "error: ". You can use direct message if that is better, or through tech support. HIH! -Miguel

M_Maldonado · ‎11-15-2015

Hi Albe, Below a quick thing to try. I have another couple of ideas depending how complex your diagram is. If this is urgent, please also submit a Tech support form to confirm what's going on. Troubleshoot What happens if you add just a Score node to the "Best" Boosting model. Just Boosting->Score, do not connect your scoring data, does that run OK? If yes, you can use the optimized score code to give it a try to manually score your test data. How Complex is your diagram? can you share a screenshot? If there are no transformations or new variables in that flow, then you might use the EMFlowscore that your gradient boosting node has already produced. Start with these two ideas and let's figure out what's happening with that node of yours. Thanks, Miguel

M_Maldonado · ‎11-12-2015

Download the Files (GitHub) This "learn by example" tip walks you through a classic credit scoring data mining flow. Use this link to download the files you need from SAS Software's github. You can import the XML into SAS Enterprise Miner to recreate a flow diagram. Use the PDF file of each example to read more details on how to build your flow diagram step by step. In a nutshell, you create the basic Credit Scoring flow for the CS_ACCEPTS data set using these nodes: Data Partition node to create a validation set and prevent overfitting Interactive Grouping node to bin your input variables, visually inspect the weight of evidence trend, and override with a manual weight of evidence if necessary. Scorecard node to train a logistic regression on the weight of evidence of your binned inputs, and to translate the results of your regression into a scorecard that is easy to interpret. As a bonus, you create a Reverse Scorecard. With one quick property change, you change the order of the scale for your scorecard. Usually the higher the score, the less likely you expect to see an event (payment default). In a reverse scorecard it is the opposite, the higher the score the more likely you expect to see an event. Another example in the same repository walks you through a flow diagram for Reject Inference. You can use use this diagram to account for sample bias, as long as you have a Rejects data set. As part of the documentation of the repository for this example, the PDF document briefly describes the fuzzy method for reject inference. Find more information about other methods in the Credit Scoring section of the Reference Help, and a visual explanation of these methods in the video, Reject Inference in SAS Enterprise Miner: Useful links: How to build a scorecard using Credit Scoring for SAS® Enterprise Miner™, is another tip that walks you through some detailed screenshots on how to manually adjust the weight of evidence of your groupings, and how to interpret your scorecard model. SAS Enterprise Miner Reference Help: SAS Credit Scoring Siddiqi, Naeem, Credit Risk Scorecards: Developing and Implementing Intelligent Credit Scoring, Cary, NC: SAS Press, 2005. Find similar learn by example tips in this directory: Learn by Example using SAS® Enterprise Miner™ and in the Data Mining GitHub repository. Feel free to post questions or comments! -Miguel

M_Maldonado · ‎11-09-2015

Hey bud, Glad you are back. I'm the same, I always use a test set. At the beginning of the your flow do: Data->Data Partition (70%training, 0% validation, 30%test)->then your crossvalidation flow->then a Model Comparison node [even if you only have one model, you are using it to see more fit statistics about your model]. Keep up the good work! -Miguel

M_Maldonado · ‎11-07-2015

Hi Frak, You are right, three nodes that support crossvalidation directly are: Decision tree LARS regression HPTree For the rest of the model nodes (including some High-Performance model nodes), you can use the Start and End group nodes. As long as the node that you are using produces SAS score code, you can use it. This means most of the nodes (HP Regression, HP Tree, HPNeural, etc) will work well with this approach, but not HPForest or some settings of HPSVM. Check out: The Power of the Group Processing Facility in SAS® Enterprise Miner™, Sascha Schubert, SAS Institute Inc., Cary, NC https://support.sas.com/resources/papers/proceedings10/123-2010.pdf This paper walks you through all the cool things you can do with the Start and End group nodes. Bottomline, you need to use a Transform node to create a crossvalidation variable. Include that, and you are pretty much setup! Good luck! Miguel [Edit: I added HPTree to the list. You can do crossvalidation cost-complexity when you do not have a Partition node]

M_Maldonado · ‎11-03-2015

The Text Parsing node identifies one or more roles for each term. For example, the term "player" on all examples we discussed would have the role of 'noun'. Think of examples of the same term with a different role. For example in "he is such a player", you would hope the parsing picks up "player" as an adjective. Maybe there are better examples 🙂 Notice that the parsing does pick up n-gram terms. On the results click on "role" to sort them by role, and try to find 'Noun Group'. Those are the best examples of n-grams. Also keep your participation high on this other SAS community. Lots of interesting stuff going on there as well: https://communities.sas.com/t5/SAS-Text-and-Content-Analytics/bd-p/text_analytics Good luck with your text!

M_Maldonado · ‎11-03-2015

Here a similar thread in case it helps. https://communities.sas.com/t5/SAS-Procedures/Proc-Logistic-Wald-Chi-Square/td-p/104021

Online Status	Offline
Date Last Visited	‎02-28-2018 11:39 AM

Re: Unbalanced data - miner

Re: SAS EM only: How to use parameter estimates in the next node?

Re: StatExplore Node

Re: How many leaves and nodes should a tree

Re: Export scoring code for Cross Validation in SAS Enterprise Miner

Re: Export scoring code for Cross Validation in SAS Enterprise Miner

Re: run time error ensemble model

Re: run time error ensemble model

Re: help with hash table

Re: help with hash table

Re: StatExplore Node

Re: How to access Variable importance in neural network in EM?

Re: Grouping variables to create new variables SAS Enterprise Miner

Re: Error when running market basket node in SAS EM

Re: Seed Initialization Method for Hierarchical Clustering

Re: Using cross-validation in Enterprise Miner;

Re: How come no Segment Profile after I set "Cluster Variable Role" = ...

Re: Confusion matrix in Enterprise Miner

Re: How can we export dataset from enterprise Miner as a csv file or t...

Re: How many leaves and nodes should a tree

Credit Scoring by Example in SAS® Enterprise Miner™

Tip: How to model a rare target using an oversample approach in SAS® ...

Tip: How to interpret your SAS® Rapid Predictive Modeler results

Tip: Use the Cutoff Node in SAS® Enterprise Miner™ to Consume the Post...

Tip: How to build a scorecard using Credit Scoring for SAS® Enterprise...

Re: E Miner - Link Analysis

Re: run time error

Re: run time error

Re: SAS EM Decision Tree English Rules

Re: SAS Enterprise Miner: Decision Tree - Test Data

Re: SAS Miner: Missing Values

Re: merde node exits with error

Re: Variables in Random Forests in SAS EM

Re: System error in scoring phase with Gradient Boosting model

Re: System error in scoring phase with Gradient Boosting model

Credit Scoring by Example in SAS® Enterprise Miner™

Re: Using cross-validation in Enterprise Miner;

Re: Using cross-validation in Enterprise Miner;

Re: text miner

Re: Proc Logistic Wald Chi-Square Calculation in EM, Please Help. Than...