Don’t get me wrong, I’m a big Python fan, both for unsupervised and supervised learning. However, since Kramp Hub’s ¹ data warehouse is based on Google Cloud Platform (GCP), an alternative solution got my attention this time: BigQuery ML.

BigQuery Machine Learning (ML) is a GCP’s feature to operationalize ML algorithms directly within the BigQuery environment. Using only SQL, models can be developed, trained, and used for predictions. With that, it democratizes ML’s operationalization for data analysts and other DWH users. Moreover, due to its integration with the DWH, models’ development speed and complexity are reduced.

“BigQuery ML brings ML to the data”.

It supports a wide range of models, as Linear and Binary Regression, Deep Neural Network, XGBoost, K-means clustering, among others. Considering that our topic is customer segmentation, can you guess the chosen one? Yes, you got it right, I’ll talk about K-means²

But first, let’s dive into the business problem that we were trying to solve.

#bigquery #data #machine-learning #python

Customer Segmentation with BigQuery ML: Unsupervised Learning Without Python!
2.90 GEEK