Google Professional Machine Learning Engineer Question 66
Single answerGoogle Cloud PlatformYour company wants to develop a machine learning model to classify customer reviews as positive, neutral, or negative. The team has limited machine learning expertise and a dataset of 50,000 labeled customer reviews. They want to ensure the model is trained and deployed quickly without requiring extensive code or manual tuning. Which Google Cloud service or feature should they use?
- A
AutoML Natural Language
- B
Vertex AI Custom Training
- C
BigQuery ML
- D
TensorFlow on Google Cloud
Show answer and explanation
Correct answer: A
Explanation
AutoML Natural Language is ideal for this scenario because it provides an automated, no-code solution for training high-quality text classification models. It is specifically designed for users with limited machine learning expertise, making it the most appropriate choice given the team's requirements for simplicity and speed.
- A. Correct.
AutoML Natural Language is the correct choice as it is specifically designed for text classification tasks and allows users with limited ML expertise to train high-quality models using a simple, no-code interface.
- B. Incorrect.
Vertex AI Custom Training is designed for users who need maximum control over model design and training, which may not be suitable for users with limited expertise or for scenarios requiring quick deployment.
- C. Incorrect.
BigQuery ML allows you to build ML models directly in BigQuery, but it is not optimized for tasks like text classification compared to AutoML Natural Language.
- D. Incorrect.
TensorFlow on Google Cloud provides the flexibility to build and train custom models, but it requires significant expertise in machine learning, which the team lacks.