AI-900 Question 120
Select 3You are tasked with building a machine learning model to classify customer reviews as positive or negative. To ensure your model generalizes well, you split your data into training and validation datasets. Which of the following statements correctly describes the purpose of these datasets?
- A
The training dataset is used to teach the model by adjusting its internal parameters.
- B
The validation dataset is used to evaluate the model's performance during training and tune hyperparameters.
- C
The training dataset is used to test the final performance of the model on unseen data.
- D
The validation dataset is used to improve the model's accuracy by directly modifying its parameters.
- E
The validation dataset helps detect overfitting during the training process.
Show answer and explanation
Correct answers: A, B, E
Explanation
In machine learning, the training dataset is used to train the model by updating its internal parameters, while the validation dataset is used to evaluate the model's performance during training, helping to tune hyperparameters and detect issues like overfitting. This ensures the model generalizes well to unseen data. The test dataset, which is separate from both, is used to evaluate the final performance of the model.
- A. Correct.
Correct. The training dataset is specifically used to teach the model by adjusting its internal parameters through the learning process.
- B. Correct.
Correct. The validation dataset is used during training to evaluate the model's performance and tune hyperparameters such as learning rate or batch size.
- C. Incorrect.
Incorrect. The training dataset is not used to test the final performance. Testing is performed on a separate test dataset, not the training dataset.
- D. Incorrect.
Incorrect. The validation dataset does not directly modify the model's parameters but instead provides feedback to adjust hyperparameters or detect issues like overfitting.
- E. Correct.
Correct. The validation dataset is used to monitor the model's performance during training and can help detect overfitting if the model performs significantly better on the training data than on the validation data.