Professional Machine Learning Engineer

Question

You work for a gaming startup with several terabytes of structured data in Cloud Storage. The data includes gameplay-time data, user metadata, and game metadata. You want to build a model that recommends new games to users while requiring the least amount of coding. What should you do?

Explanation

BigQuery ML matrix factorization is designed for collaborative-filtering recommendation systems and can use historical user–game interactions, such as gameplay time, as implicit feedback. Loading the structured data into BigQuery and training the model with BigQuery ML avoids the custom TensorFlow implementation required by notebook-based two-tower or matrix-factorization approaches. The CREATE MODEL statement for matrix factorization models

Learn more

Professional Machine Learning Engineer

AstroTutor

Notes

Readiness

QuestionQ40

Community Discussion