D491 Introduction to Analytics - Set 5 - Part 1

Correct Answer: Feature importance scoring

Random forests use feature importance scoring to determine which variables contribute most to the model's decision-making process.

Correct Answer: Supervised learning

Supervised learning uses labeled data to train the model to make predictions or classifications.

Correct Answer: Decision trees can handle categorical variables without encoding

Decision trees can directly handle categorical variables without needing one-hot encoding, unlike linear regression.

Correct Answer: To determine the optimal number of clusters

The elbow method helps identify the optimal number of clusters in K-means by plotting the explained variance and finding the "elbow" point.

Correct Answer: Confusion matrix

A confusion matrix is commonly used to evaluate the performance of a classification model by comparing predicted and actual class labels.

Correct Answer: F1 score

The F1 score balances precision and recall, making it ideal for evaluating the performance of binary classification models.

Correct Answer: Principal Component Analysis (PCA)

PCA is commonly used for dimensionality reduction by transforming high-dimensional data into a smaller set of principal components.

Correct Answer: To create new input features that improve model performance

Feature engineering involves creating new features that improve the model's predictive accuracy or efficiency.

Correct Answer: Convolutional Neural Networks (CNNs)

CNNs are highly effective for image classification due to their ability to capture spatial hierarchies in images.

Correct Answer: To calculate the gradient of the loss function and update weights

Backpropagation calculates the gradient of the loss function and uses it to update the weights of the neural network during training.

Correct Answer: Overfitting

Regularization techniques, such as L1 and L2, are used to prevent overfitting by adding penalties to complex models.

Correct Answer: Ridge regression

Ridge regression (L2 regularization) is used to handle multicollinearity by adding a penalty to the regression coefficients, thus shrinking them.

Correct Answer: To split text into individual words or phrases

Tokenization involves breaking down a string of text into individual tokens, such as words or phrases, for easier analysis.

Correct Answer: Mean Squared Error (MSE)

MSE is commonly used in regression tasks to measure the average squared difference between predicted and actual values.

Correct Answer: KNN has a high computational cost

KNN has a high computational cost because it requires calculating the distance between the input and every point in the dataset during prediction.

Correct Answer: Apriori

The Apriori algorithm is used in market basket analysis to identify frequent itemsets and generate association rules.

Correct Answer: To scale numerical features to a common range

Normalization scales numerical features to a common range, improving the performance of algorithms sensitive to feature scale, such as K-means and neural networks.

Correct Answer: They use decision trees as base models

Gradient boosting algorithms use decision trees as base models and iteratively improve model performance by reducing the residual error.

Correct Answer: To control the step size for updating weights during training

The learning rate controls how much the model's weights are updated during each iteration of training.

Correct Answer: The model incorrectly predicted a positive class as negative

A false negative occurs when the model incorrectly predicts a positive class instance as negative.