Creating Prediction

Application Name: This is the kind of business area where prediction needs to be applied. The options available are CRM (Customer Relation Management), WFA ( Workforce Analytics) and Generic.
Prediction Name: Give a name to your prediction.
Call Type: There are two options here:

Synchronous: Selecting this option will run multiple predictions simultaneously.
Asynchronous: Selecting this option will run one single prediction process at one time, the next process will run after the current process is completed.

Algorithm: There are multiple pre-built algorithms that can be used to run the prediction process as per need.
Algorithm Params: This field will be auto filled with the algorithm parameters when a algorithm is selected.
Model Restore: Selecting Yes will store the current prediction model for further reference and usage.

Input Type: Select the type of data input you would be providing. There are two options here: Table and File. This data should be same for the output.
Select the database connection from the drop down provided. This data should be same for the output.
Translation Table: For Prediction to work, the entire table content should be in numerical format, for that purpose a translation table is used. Enter the table name here. This data should be same for the output.
Primary Key Column: This is the primary key in the table. This data should be same for the output.
Training Dataset Query: For every prediction to work properly, the system is fed with proven data that is used as a reference for any prediction to take place. Training Dataset Query represents that proven data from which inferences are taken. This constitutes for 80% of the data
Test Dataset Query: This is the query for the prediction is executed. This constitutes for 20% of the data.

In the next step, you can select the factors on which you need predictions to happen.

You have all the available factors from which you can move them to X-axis (Selected Factors), and Y-axis (Selected Labels). In the selected labels section there are two options: Dependent and Multiple.
Select multiple if there are more than two entries used for prediction. Select Dependent for at most 3 entries. Dependent will execute the prediction query for the first entry and then depending on the first entry result, the second entry is executed.
Click Configure Influential Factors. This option displays a pop-up where the influence percentage of a particular factor is displayed. Select the factors and click Use Selected Factors.