Creating a New Predictor

Businesses often need to identify the likelihood of future outcomes based on historical data. The goal is to go beyond knowing what has happened to provide the best assessment of what will happen in the future. For example, using historical data, a model can be trained to detect and avoid fraud transactions, optimize marketing strategies by predicting customer responses or purchases, and improve operations by predicting inventory or required resources. 

To create a new predictor, you need dataset. So there are two steps in creating a new Predictor:

• Preparing Datasets
• Creating a new Predictor Model

Preparing Datasets:
A custom predictor can easily be built using Smart Bot with a .csv (or .xlsx) dataset containing multiple columns. The .csv dataset is a collection of historical records with multiple columns. The last column should be the target column which you want to predict which depends on other preceding columns. For example, in the following sample dataset, you want to predict whether the health insurance will be claimed or not, which is the last column. It depends on other independent variables such as the applicant's age, sex, bmi, children (no. of dependents), smoking habit, region, and insurance charges.

Model page helps to select uploaded datasets from which you can train the model.

Creating a new Predictor Models:

1. Navigate to Smart Bot > Predictive Analysis.

The Predictive Analysis page displays two tabs namely Dataset and Model.

2. Open the Model page.
3. Click Add Model.  

The Classifier Configuration window is displayed.

1. Select Create.
2. Enter the unique Predictor Name. It should start with a letter and can contain only letters, numbers, and underscore.
3. Select the required dataset from the Dataset drop-down.
4. Select Method option from the drop-down. The supported methods are:
   • RF
   •  XGB
   • ANN

This selection is optional and are for advanced user. They can select any method to train and compare results of multiple models train. The default method is RF.

5. Optionally enter additional Description.
6. Click Submit.

A model is created in the list with the status ‘In Progress’. The Smart Bot will take some time to train the recognizer based on the size of the dataset. When the training is completed, Smart Bot will update the status to 'Completed'.