A SQL-Based Probabilistic Inference
Feng-Jen Yang · 2023
There is an industrial trend to integrate machine learning models into relational databases so that users can get the intended support without compiling and feeding the set of training data to a learning model and training it separately. To this end, implementing a machine learning model by using SQL statements is an efficient and low-cost practice because the relational database platform is ready to run the intended computational queries without additional plugins. By integrating probabilistic inference functionalities into relational databases, we can ensure that decision-makers have access to the most accurate and up-to-date information possible. This is especially important in today's fast-paced world where the ability to make quick and informed decisions is more important than ever. In this study, I demonstrated a way to equip a relational database with the capability of performing probabilistic inference by using only SQL statements. This is an exciting development that has the potential to reshape the way we design databases and approach decision-making in the industrial world.