A new spatial enhancement in Oracle Machine Learning enables organizations to include location relationships in their ML models for improved model accuracy. Developers and data scientists can now create machine learning models incorporating location relationships and location-based predictive analysis at scale.

Why does this matter to you?

Data scientists may try to perform such specialized modeling locally; however, that requires moving massive amounts of data, writing complex algorithms, and managing the development environment independently. In contrast, the new Spatial enhancement to Oracle Machine Learning for Python (OML4Py) allows developers to detect spatial patterns through quantitative approaches—such as spatial clustering, regression, classification, and anomaly—without moving the data outside the database. For example, you can now build an ML model that may recognize that transformers in coastal areas are more sensitive to age and moisture, where saltwater effects accelerate degradation.