GeeCON 2023: Marton Kodok - Discover BigQuery ML, build your own make MODEL stmt

With BigQuery ML, you can build device learning models without leaving the database environment and training it on massive datasets. In this demo session we are going to show common marketing device Learning usage cases of how to build, train, eval, and predict, your own scalable device learning models utilizing SQL language in Google BigQuery and to address the following usage cases: - client Segmentation + Product cross sale advice - Conversion/Purchase prediction - Inference with another in-built )20 models The audience will get first-hand experience with how to compose make MODEL sql syntax to build device learning models specified as: - Multiclass logistic regression for classification - K-means clustering - Matrix factorization - ARIMA time series predictions ... and more Models are trained and accessed in BigQuery utilizing SQL — a language data analysts know. This enables business decision-making through predictive analytics across the organization without leaving the query editor. In the end, the audience will learn how everyday developers can build/train/run their own machine-learning models consecutive from the database query editor, by issuing make MODEL statements