Workflow Automation Templates
A library of ready-to-use workflow templates to accelerate your data journey
Classification Model Building Label Data
Build and apply classification models
Overview
This workflow automates the process of building a classification model using H2O’s Gradient Boosting Machine (GBM) to analyze transactional or labeled data. It performs data preprocessing, model training, scoring, and result generation, enabling accurate anomaly detection or categorical outcome prediction.
Details
The workflow begins by importing labeled data and applying conditional transformations using the Case When node to create or refine target variables. The processed data is then used to train an H2O Gradient Boosting Machine, a powerful ensemble learning method that combines multiple weak learners to deliver high prediction accuracy.
Once trained, the model is applied to new data through the H2O Score node, generating predictions and associated probabilities. The output is further refined using steps like Print N Rows for previewing results, Drop Columns for retaining essential fields, and Save CSV for exporting final predictions.
This end-to-end workflow supports explainable AI with SHAP-based interpretability, allowing users to understand the impact of each feature on predictions. It is ideal for classification tasks such as fraud detection, customer segmentation, or risk assessment.