This use case aims to predict the median income of certain regions in the city of Los Angeles according to certain features from the Los Angeles Income Census dataset.

The dataset is stored in the la_block_groups table in the database. Based on spatial analysis and statistics, the example trains a regression model to estimate the median income.

During the exercise, different statistics are calculated to find out which spatial model is more suitable for the given data. AdaptiveSpatialRegressor internally computes all those statistics and suggests the best model. However, to show the analysis capabilities of Oracle Spatial AI, the exercise goes through different steps to find the best regression model.

The following steps enable you to solve a regression task using an OML notebook.