GMLE別格な問題集をダウンロードして無料で最新の（GMLEテスト問題集をゲット2025年02月27日)
GMLE問題集は合格保証します合格できるGMLE試験問題2025年更新
質問 # 38
Which of the following are important considerations when acquiring data from SQL databases?
(Choose two)
Response:
- A. Ignoring null values to prevent any potential issues
- B. Optimizing queries to handle large datasets efficiently
- C. Ensuring data integrity by maintaining relationships between tables
- D. Using raw, unvalidated data for better model accuracy
正解：B、C
質問 # 39
What does 'one-hot encoding' do in the preprocessing of categorical data?
Response:
- A. It reduces the dimensionality of the data
- B. It scales all features to a uniform range
- C. It identifies and removes outliers
- D. It converts categorical variables into binary vectors
正解：D
質問 # 40
What role does 'batch size' play in training neural networks?
Response:
- A. It adjusts the learning rate of the model
- B. It determines the maximum number of features used in the model
- C. It specifies the number of training examples used in one iteration
- D. It sets the total number of training iterations
正解：C
質問 # 41
What does 'grid search' refer to in hyperparameter tuning?
Response:
- A. A technique to search exhaustively through a specified subset of hyperparameters
- B. A method to visualize the performance of different models
- C. The process of randomly selecting hyperparameters for testing
- D. A strategy for reducing the number of hyperparameters
正解：A
質問 # 42
In machine learning, what is a 'Support Vector Machine' (SVM) primarily used for?
Response:
- A. Dimensionality reduction and feature extraction
- B. Classification and regression tasks
- C. Clustering and grouping similar data points
- D. Time-series forecasting and sequence prediction
正解：B
質問 # 43
In machine learning, what is 'model validation'?
Response:
- A. The method of improving the model's performance
- B. The technique of testing the model on unseen data
- C. The process of training the model
- D. The practice of combining multiple models
正解：B
質問 # 44
What is the main use of the NumPy library in Python for machine learning?
Response:
- A. Text processing
- B. Handling large arrays and matrices
- C. Web scraping
- D. Data visualization
正解：B
質問 # 45
Your cybersecurity team is tasked with detecting anomalies in network traffic that may indicate malicious activity. You decide to use an autoencoder for this task. After training the autoencoder on normal network traffic data, you notice that it is not accurately detecting anomalies.
What are the next steps you should take to improve the performance of the autoencoder?
Response:
- A. Train the autoencoder solely on anomalous data to improve its accuracy
- B. Adjust the size of the latent space and apply regularization techniques to reduce overfitting
- C. Retrain the autoencoder with fewer data points and remove regularization techniques
- D. Increase the complexity of the network by adding more layers and disabling dropout
正解：B
質問 # 46
In deep learning, what is a 'neural network layer'?
Response:
- A. The final output of the neural network
- B. A set of algorithms for optimizing the model
- C. A collection of neurons that process a set of input data
- D. The initial input data for the neural network
正解：C
質問 # 47
What are correlation matrices useful for in data exploration and visualization?
Response:
- A. Understanding relationships between pairs of variables
- B. Visualizing geographical data
- C. Determining the accuracy of a model
- D. Identifying missing data
正解：A
質問 # 48
What does the term 'overfitting' refer to in deep learning?
Response:
- A. A model that performs well on both training and test data
- B. A model that performs well on training data but poorly on unseen data
- C. A model that is too simple to capture the underlying pattern
- D. The process of selecting the right hyperparameters
正解：B
質問 # 49
Which is not a typical use case for CNNs?
Response:
- A. Facial recognition
- B. Object detection
- C. Image classification
- D. Natural language processing
正解：D
質問 # 50
You are tasked with building a classification model to predict whether customers will purchase a product based on their browsing behavior. You decide to use a Random Forest classifier. After training the model, you notice that the training accuracy is very high, but the validation accuracy is much lower.
What might be causing this, and how can you improve the model's generalization?
Response:
- A. The model is underfitting the data; increasing the number of features and trees will improve performance
- B. The model is likely overfitting to the training data; applying cross-validation and reducing the number of trees in the forest or using regularization techniques can help improve the generalization to unseen data
- C. The training data is too large; reducing the training set size will improve the model's performance
- D. The features are not useful; removing all but one feature will reduce overfitting
正解：B
質問 # 51
What is the main challenge in determining the optimal number of clusters for a k-means algorithm?
Response:
- A. It requires a predefined number of clusters, which may not reflect the true structure of the data
- B. It needs feature scaling before clustering
- C. It increases the computational complexity
- D. It depends on the amount of training data
正解：A
質問 # 52
Which of the following are common activation functions used in neural networks?
(Choose two)
Response:
- A. ReLU (Rectified Linear Unit)
- B. Batch normalization
- C. Weight initialization
- D. Sigmoid function
正解：A、D
質問 # 53
What refers to models capturing noise in the training data as if it were a true signal in supervised learning models?
Response:
- A. Models performing equally on training and test data
- B. Overfitting
- C. The process of training models on large datasets
- D. Models that are too simplistic to capture underlying patterns
正解：B
質問 # 54
In CNNs, what is the purpose of pooling layers?
Response:
- A. To increase the depth of the network
- B. To increase the network's sensitivity to noise
- C. To reduce the spatial size of the representation
- D. To directly classify images
正解：C
質問 # 55
Stochastic Gradient Descent differs from traditional Gradient Descent by:
Response:
- A. Updating model parameters after evaluating the entire dataset
- B. Updating model parameters after evaluating each data point
- C. Eliminating the need for a learning rate
- D. Using a fixed learning rate throughout the training process
正解：B
質問 # 56
What does Bayes' Theorem provide in the context of machine learning?
Response:
- A. A mechanism for speeding up computations
- B. A method for calculating the likelihood of different hypotheses
- C. A strategy for optimizing neural networks
- D. A technique for visualizing data distributions
正解：B
質問 # 57
Which metric is commonly used to evaluate the performance of a classification model?
Response:
- A. Root Mean Squared Error (RMSE)
- B. Mean Absolute Error (MAE)
- C. R-squared
- D. Accuracy
正解：D
質問 # 58
What does the term 'epoch' refer to in the context of training a deep learning model?
Response:
- A. The process of tuning hyperparameters
- B. A single iteration over the entire dataset
- C. The final phase of model training
- D. The initial phase of model training
正解：B
質問 # 59
In a CNN, what is the purpose of the pooling layer?
Response:
- A. To increase the resolution of the input image
- B. To normalize the input data
- C. To combine features and reduce the dimensionality of the data
- D. To add non-linearity to the model
正解：C
質問 # 60
Which of the following is the median of the dataset {3, 7, 9, 12, 15}?
Response:
- A. 0
- B. 1
- C. 2
- D. 3
正解：C
質問 # 61
......
検証済みのGMLE問題集で問題と解答で合格保証試験問題集テストエンジン：https://jp.fast2test.com/GMLE-premium-file.html