Using DataAnalysis

This time, instead of using Discovery to view the data report, we will use a transformation called DataAnalysis, which is located in the AI section:

We add this transformation and connect it to the Union of the data from all the DataWrangler boxes created in the previous step. To view the final analysis results, we connect our step to a Crossdata-type output for storage in our virtualizer, resulting in the following structure:

Next, we configure our DataAnalysis transformation. We specify that we want to use the Riesgo column as the object of our analysis and set the analysis type to "ML exploratory analysis," leaving all other parameters as default:

Remember to configure the Writer section to define the table name where the analysis data will be stored (e.g., analisis_datos_USERNAME). If we open the debug console:

By selecting the DataAnalysis transformation, we can see the results of this step applied to a data sample:

Finally, we execute our workflow and use the Rocket catalog in our project to run a query and check the final result:

SELECT * FROM default.analisis_datos_USERNAME

The result should look something like this:

{"frequencies":[X,Y,Z],"values":["amarillo","rojo","verde"]}

We compare the values X, Y, and Z obtained here with those displayed in the Discovery graph from the previous step. Finally, we document the X, Y, and Z values obtained as part of our deliverables.