D491 Introduction to Analytics - Set 3 - Part 2

Correct Answer: Principal component analysis (PCA)

PCA is commonly used to reduce the number of variables in a dataset by transforming them into a smaller set of uncorrelated components.

Correct Answer: Imputation

Imputation is the process of replacing missing data with substituted values to maintain dataset integrity for analysis.

Correct Answer: Linear regression

Linear regression is used to predict continuous numerical outcomes based on the linear relationship between dependent and independent variables.

Correct Answer: To estimate the performance of a model

Cross-validation is used to assess how well a machine learning model will generalize to an independent dataset by splitting data into multiple training and validation sets.

Correct Answer: Silhouette coefficient

The silhouette coefficient measures how similar objects are within a cluster compared to objects in other clusters, helping to evaluate clustering quality.

Correct Answer: To display the actual vs. predicted class distribution

A confusion matrix displays the true positives, true negatives, false positives, and false negatives, helping evaluate classification model performance.

Correct Answer: Isolation forest

Isolation forests are used for detecting anomalies by isolating data points that behave differently from the majority of the dataset.

Correct Answer: Scatter plot

A scatter plot displays the relationship between two continuous variables, helping visualize any correlation.

Correct Answer: It minimizes variance within clusters

K-means clustering works by minimizing the variance within each cluster, making data points within a cluster more similar to each other than to those in other clusters.

Correct Answer: Pruning

Pruning helps reduce overfitting in decision trees by trimming branches that provide little predictive power.

Correct Answer: Association rules

Association rules are used to discover patterns or relationships between variables in large datasets, such as frequently bought items in market basket analysis.

Correct Answer: It reduces the number of features in the model

Lasso regression applies regularization that shrinks some feature coefficients to zero, effectively reducing the number of variables used in the model.

Correct Answer: To standardize the range of independent variables

Feature scaling standardizes the range of features so they contribute equally to the model's performance, especially for algorithms sensitive to varying scales like SVM or K-means.

Correct Answer: Oversampling the minority class

Oversampling the minority class involves replicating the smaller class to balance the dataset and improve model performance for imbalanced classification tasks.

Correct Answer: To indicate the outcome or predicted class

Leaf nodes in a decision tree represent the final predicted class or outcome after following the decision rules along the branches.

Correct Answer: Convolutional Neural Network (CNN)

CNNs are commonly used for image classification because they can efficiently capture spatial hierarchies in images.

Correct Answer: Root mean squared error (RMSE)

RMSE measures the average difference between the actual and predicted values, helping evaluate the accuracy of a regression model.

Correct Answer: To prevent overfitting

Regularization techniques like Lasso and Ridge prevent overfitting by adding a penalty to the size of the coefficients, encouraging simpler models.

Correct Answer: The proportion of true positives among all predicted positives

Precision measures the proportion of correctly predicted positive observations out of all predicted positives, assessing a models performance in classification.

Correct Answer: To show correlations between variables

Heat maps are used to display correlations or relationships between variables by using color intensity to indicate values.