Data Preprocessing for Deep Learning QUIZ (MCQ QUESTIONS AND ANSWERS)

Minimize classification error

Maximize accuracy

Minimize the mean squared error between predictions and actual values

Maximize precision and recall

To initialize the weights of the neurons

To compute the gradient during backpropagation

To introduce non-linearity into the model

To determine the learning rate

Support Vector Machine (SVM)

K-Means Clustering

Decision Tree

Principal Component Analysis (PCA)

The model performs well on unseen data

The model fails to capture the underlying patterns in data

The model fits the training data too closely, capturing noise

The model has too few parameters

Increasing data complexity

Reducing the dimensionality of data

Preprocessing raw data

Enhancing model training time

To increase the complexity of the dataset.

To create new features for the model.

To clean and transform raw data into a suitable format for model training.

To reduce the number of samples in the dataset.

Principal Component Analysis (PCA)

T-test

Data augmentation

Data whitening

It increases the model's complexity.

It may lead to overfitting.

It removes important data points.

It improves model generalization.

It simplifies the model architecture.

It improves model training speed.

It ensures that all classes have an equal number of samples.

It prevents the model from being biased toward the majority class.

To confuse the model.

To remove categorical variables from the dataset.

To convert categorical data into a numerical format.

To create additional features.

By increasing the number of features.

By ignoring the issue; it does not affect models.

By removing features that are highly correlated.

By converting all features to categorical data.

Enhancing the accuracy of models

Handling imbalanced data

Increasing the complexity of the data

Improving the robustness of the model

To create new data samples.

To split the dataset into training and testing sets.

To ensure the model generalizes well and avoid overfitting.

To transform categorical data into numerical values.

Replacing words with synonyms

Adding background noise

Flipping horizontally or vertically

Applying PCA

Under-sampling

Over-sampling

Dimensionality Reduction

SMOTE

Convert timestamps to categorical data.

Convert timestamps to numerical features.

Ignore timestamps during preprocessing.

Use timestamps as labels for the model.

It decreases the model's complexity.

It reduces the size of the dataset.

It increases the amount of available training data.

It removes noise from the data.

Deleting rows with missing values.

Replacing missing values with zeros.

Ignoring missing values during training.

All of the above.

To make data more colorful.

To convert text data into numerical data.

To decorrelate features and normalize data.

To increase the dimensionality of the data.

To increase the model's complexity.

To change the feature names.

To ensure all features have the same scale.

To add noise to the data.

Before data cleaning.

After data normalization.

When dealing with categorical data.

Only for time-series data.

It improves model training speed.

It simplifies the model architecture.

It ensures that all classes have an equal number of samples.

It prevents the model from being biased toward the majority class.

To increase the number of features.

To remove features with low variance.

To reduce the number of features while preserving relevant information.

To transform categorical data into numerical data.

To convert text into images.

To convert text into numerical data.

To increase the complexity of text data.

To remove stopwords from text.

Removing the majority class.

Reducing the learning rate.

Oversampling the minority class.

Increasing the number of training epochs.

It reduces the dataset size.

It increases the model's complexity.

It helps the model generalize better.

It makes the model overfit the data.

Principal Component Analysis (PCA)

Data augmentation

Data scaling

Data whitening

It reduces the dimensionality of the data.

It improves model interpretability.

It converts categorical data into a numerical format.

It increases the complexity of the data.

To finalize the model architecture.

To visualize data for better understanding.

To increase the model's complexity.

To shuffle the data randomly.

To increase the noise in the data

To make the data more complex

To scale data to a specific range

To remove outliers from the data