Define ML Models

Infoveave ML Modeller is a powerful tool designed to facilitate the creation of machine learning models. It streamlines the process of creating, modifying, and refining machine learning models, empowering you to make data-driven decisions effectively.

Steps to Create and Setup Machine Learning (ML) Model

Access Machine Learning (ML) Model

1. To create and execute a What-If analysis click on Analysis ML Modeller.You can find all your ML models under the tabs My ML Models and Shared ML Models respectively.
2. To create new model, click on New ML Model. It will redirect you to the ML model builder.
3. Provide a meaningful Name in the Setup panel. This will help you to identify and understand the purpose of each ML model.
4. Click on Save to save the model.

Select & Validate Query

The query activity sets the base foundation for your machine learning (ML) model. By validating the query, you gain the ability to perform various data transformation activities essential for refining and preparing your data for ML modelling.

1. Drag and drop the Query activity from the Activity panel to the ML model builder.
2. Click on the Query activity to select the desired query from the dropdown list.For information on creating queries, refer to the Queries.
3. Click on the Validate and Preview button to validate the selected query.Once the query is validated, it will display the data table.Validating allows configuring more data transformation activities and in-built ML models.

Tip

The green check box appears on each activity after successful activity validation.

Add & Validate Data Transformation Activities

Data transformation activities helps you changing and organizing your data ready for machine learning analysis. It also ensures that the data is in the right format and structure for effective modeling.

1. Drag and drop all the other required data transformation activities from the Activity panel to the ML model builder.
2. Select the activity and configure them individually in the Configuration panel.
3. Click on Validate and Preview button to validate each activities.
4. Link the activities with other configured activities as required.
5. Link/chain all the configured data transformation activities in the ML model builder.

Note

You must validate each individual activities in the ML model separately. The process of configuring each data transformation activities varies based on the activities type.

Tip

The green check box appears on each activities after successful activities validation.

Select & Build ML Model

The query activities sets the base foundation for your machine learning (ML) model. By validating the query, you gain the ability to perform various data transformation activities essential for refining and preparing your data for ML modelling.

1. Drag and drop the required ML model from the Activity panel to the ML model builder.
2. Select the ML model and configure the model in the Configuration panel.
3. Link/chain all the configured data transformation activities with the configured ML model in the ML model builder.
4. Click on Validate and Build button, to validate and build the ML model in Infoveave.
5. It will start building the model. After the successful build of the model, the model will give the details on the following
   • Area under ROC curve
   • Accuracy
   • Positive precision
   • Positive recall
   • Negative precision
   • Negative recall
   • F1 score
   • Area under precision recall curve
   • Confusion matrix
6. Select the model with the highest accuracy to predict the outcome with more input data.

Note

You must link/chain all the configured activities in the ML model builder to Validate and Build the ML model. You can use only one ML model to train and build the model.

Tip

The green check box appears on each activities after successful activities validation.