Predicting churn probability using labeled past outcomes aligns with which AI approach?

Predicting churn probability using labeled past outcomes aligns with which AI approach?

• A. Rule-Based
• B. Analytical AI
• C. Generative AI
• D. Hybrid AI

Answer: (B)

The main idea here is using historical, labeled outcomes to forecast a future probability. When you have past records of whether customers churned and the associated features, you can train a model to map those features to the likelihood of churn for new customers. This data-driven, probabilistic forecasting is at the heart of Analytical AI, which focuses on extracting insights from data and making predictions. Rule-based approaches rely on hand-crafted rules set by humans and don’t learn from data in the same way, so they aren’t the best fit for forecasting churn from historical outcomes. Generative AI is about creating new content rather than predicting outcomes from past data. Hybrid AI could mix methods, but the scenario described aligns most directly with trained predictive analytics, i.e., Analytical AI.

The main idea here is using historical, labeled outcomes to forecast a future probability. When you have past records of whether customers churned and the associated features, you can train a model to map those features to the likelihood of churn for new customers. This data-driven, probabilistic forecasting is at the heart of Analytical AI, which focuses on extracting insights from data and making predictions.

Rule-based approaches rely on hand-crafted rules set by humans and don’t learn from data in the same way, so they aren’t the best fit for forecasting churn from historical outcomes. Generative AI is about creating new content rather than predicting outcomes from past data. Hybrid AI could mix methods, but the scenario described aligns most directly with trained predictive analytics, i.e., Analytical AI.