PMLE Architecting Low-Code ML Solutions Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  CREATE MODEL `project.dataset.recommender` OPTIONS(model_type='kmeans', num_clusters=10) AS SELECT user_id, product_id, rating FROM `project.dataset.interactions`

  Why it's wrong here

  This statement creates a k-means clustering model, which is used for unsupervised clustering, not for recommendation. K-means does not leverage user-item interactions to make recommendations. The company needs a collaborative filtering approach, which matrix factorization provides. Therefore, this statement does not meet the requirements.

• ✗

  CREATE MODEL `project.dataset.recommender` OPTIONS(model_type='logistic_reg', user_col='user_id', item_col='product_id', rating_col='rating') AS SELECT user_id, product_id, rating FROM `project.dataset.interactions`

  Why it's wrong here

  This statement attempts to create a logistic regression model with options meant for matrix factorization. Logistic regression is used for binary classification, not for collaborative filtering. The options user_col, item_col, and rating_col are not valid for logistic regression. Therefore, this statement will fail to create a proper recommendation model.

• ✓

  CREATE MODEL `project.dataset.recommender` OPTIONS(model_type='matrix_factorization', user_col='user_id', item_col='product_id', rating_col='rating') AS SELECT user_id, product_id, rating FROM `project.dataset.interactions`

  Why this is correct

  This statement correctly creates a matrix factorization model in BigQuery ML. It specifies the model type as matrix_factorization and identifies the user column, item column, and rating column. The SELECT statement provides the training data with user-item interactions and ratings. This is the standard syntax for building a recommendation model using collaborative filtering in BigQuery ML.

• ✗

  CREATE MODEL `project.dataset.recommender` OPTIONS(model_type='matrix_factorization', input_label_cols=['rating']) AS SELECT user_id, product_id, rating FROM `project.dataset.interactions`

  Why it's wrong here

  This statement uses the model type matrix_factorization but incorrectly specifies input_label_cols instead of the required user_col, item_col, and rating_col options. Matrix factorization requires those specific options to identify the user and item columns. Without them, the model cannot be trained correctly. Thus, this statement is incorrect.

Go deeper

Related to this question