simRegression Input/Output

simRegression can process any file that meets the Platform File Specifications as long as it has an “ID” column, which assigns a unique value to each row, and a “Regression” column, which tells the engine the column it will attempt to predict.

Using simRegression you can query your existing data to predict the Score of the Regression column. After you have filled in the information for the Query Object, the Query will return the predicted result for the Regression column along with the weighted factors behind that prediction.

simRegression Sample Output

Prediction

View the prediction of the Query under the Prediction tab. The Target column lists the object selected for the query and the Score column predicts the value of the regression target. In the example below, the goal was to predict the cost of health insurance. The Target was "charges" and the Score is the prediction of that, the predicted charges for the insurance. 

Why

View the values that determined the outcome under the Why tab. The values are presented in a weighted list to show which factors have a higher impact on the selected query object.

Query object is more similar (+)

The plus sign (+) denotes the presence of a certain attribute in the query object. The presence of this attribute is a key consideration in the model's analysis, but it does not inherently suggest a positive or negative impact on the predicted value. It signifies that the attribute is a relevant factor in the model's calculations.

Query object is less similar (-)

A negative indicator (-) represents the absence of a specific attribute in the query object, which is also taken into account in the model's prediction.