
Hello Friends,
Welcome back to my other blog post on an exciting feature available in SAP Datasphere, i.e., the New Intelligent Look Up in SAP Datasphere. In this blog post, we will understand this concept with an example.
In this blog post, I am not going to discuss what an intelligent look-up is, as there are many blog posts that talk about intelligent look-up. We will create a demo example that will help you to understand and implement the same in your project.
Now Get Set go: -
Before starting this post, I expect you had reviewed my previous post on "SAP Data Sphere Business Case Scenarios & Implementation Strategies".
If not, then kindly visit the link:
Now Let us get straight to an example :-
Now let us consider the dataset we have. We have a purchase dataset, which is our input, and we have a vendor dataset, which is our look-up.
Purchase Dataset:
Vendor Dataset:
So my input and look-up datasets are ready. So what am I going to do with these datasets? I am going to return Vendor_ID from the Vendor dataset by matching columns with the purchase dataset with the help of Intelligent Look Up.
The matching columns that I am going to use are First_Name, email, and location between the 2 datasets.
Let us see the datasets in SAP Datasphere with complete flow on the intelligent lookup created.
Now let us discuss the entire flow step by step. The moment you create an intelligent look up it is created as below.
Now, drag and drop the input file and the look-up file. In our case, it is the purchase dataset and the vendor dataset.
It looks like this, as in below.
Now let us go to Rule 1, which is connected to the datasets, and understand that.
run the intelligent look up flow and see the result.
Now let us click the Rule1 (Rule_FirstMatch) and apply more rules.
we have 2 options in a rule.They are
1. Add rule for multiple Matches 2. Add rule for Unmatched records
Here in our case we have unmacthed records hence we will add rule for matched records. Please see below image.
we can add rule for multiple matches also to rule out which is the exact and correct match. in our case we have added 1 more rule for unmatched records. please see the below image with comments.
now after running the intelligent look up flow please see the below result in the image with comments.
now, we have applied a new rule, and this time we gave the match strategy as fuzzy match.
Now we have 8% match records and we have achieved the final result. we have found all the vendor_ids from the look up file.
In case if we did not find an exact match or say 1 or 2 records are still in Unmatched state then the data stewardship person can manually approve and reject the matched records.
Intellignet look up is an powerful feature in SAP Dataspehere. This concept is taken mainly from SAP BODS ETL tool where we have look up transformation and Match Transformation. These 2 concepts are combined and added as a feature in SAP dataspehere.
This intelligent look up feature will help data modellers as well as business people to rule out the exact match when doing look up with external datasets.
Thank you.
Narasingha
SAP Data & Analytics Expert.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
10 | |
5 | |
4 | |
4 | |
4 | |
3 | |
3 | |
3 | |
3 | |
3 |