← Back to categories

Classical Machine Learning

Problem Framing & Assumptions

7 questions

Multiple Choice

Target Leakage Detection

Multi-label vs Multi-class Problem

Regression vs Classification Framing

IID Assumption Violation

Open-Ended

Choosing Between Classification, Regression, and Ranking

Understanding and Preventing Target Leakage

ML Assumptions and When They Break

Data Preparation & Feature Engineering

8 questions

Multiple Choice

Feature Scaling Timing

Handling Missing Values Strategy

Categorical Encoding for High Cardinality

Feature Engineering for DateTime

Open-Ended

When and How to Scale Features

Strategies for Handling Missing Data

Encoding Categorical Variables

Feature Selection Techniques

Models & Algorithms

14 questions

Multiple Choice

Choosing Between Linear and Tree Models

Handling Missing Values in Tree vs Linear Models

Feature Scaling Requirements

Naive Bayes Independence Assumption

Open-Ended

Understanding Linear Regression

Understanding Logistic Regression

Understanding k-Nearest Neighbors (kNN)

Understanding Decision Trees

Understanding Random Forests

Understanding Naive Bayes

Understanding XGBoost

Understanding LightGBM

Understanding CatBoost

Understanding Support Vector Machines (SVM)

Bias–Variance & Generalization

9 questions

Multiple Choice

Identifying High Bias

Regularization Techniques

Diagnosing Overfitting from Learning Curves

Cross-Validation Fold Strategy

Open-Ended

The Bias–Variance Tradeoff

Overfitting vs Underfitting

Regularization Techniques

Controlling Model Complexity

Cross-Validation Strategies

Evaluation & Metrics

8 questions

Multiple Choice

MAE vs RMSE Selection

Choosing Between Precision and Recall

Confusion Matrix Analysis

ROC AUC Limitations

Open-Ended

Reading and Interpreting Confusion Matrices

Classification Metrics: Precision, Recall, F1, and AUC-ROC

Regression Metrics: MSE, MAE, R-squared, and MAPE

Choosing the Right Metric for Your Problem

Data Splitting & Validation

7 questions

Multiple Choice

Validation Set vs Cross-Validation

Data Leakage Through Splitting

Test Set Reuse Problem

Grouped Data Splitting

Open-Ended

When and Why to Use Stratified Splits

Proper Train/Test Splitting Strategies

Time-Based Validation for Time Series

Imbalanced Learning

7 questions

Multiple Choice

Threshold Selection for Imbalanced Classes

Oversampling Timing

Evaluation Metrics for Imbalanced Data

Class Weights vs Resampling

Open-Ended

Impact of Class Imbalance on Model Training

Resampling Techniques: SMOTE, Undersampling, and Oversampling

Adjusting Decision Thresholds for Imbalanced Classification

Interpretability & Debugging

7 questions

Multiple Choice

Local vs Global Interpretability

Debugging Unexpected Model Behavior

SHAP vs Permutation Importance

Partial Dependence Plots Interpretation

Open-Ended

Methods for Feature Importance Analysis

Diagnosing Model Failures

Model-Agnostic Interpretability: SHAP and LIME

Baselines & Sanity Checks

7 questions

Multiple Choice

Appropriate Baseline for Regression

Sanity Check for Feature Importance

Single Feature Baseline

Random Permutation Test

Open-Ended

Why and How to Establish Baselines

Essential Sanity Checks Before Deployment

When Random Baseline Beats Your Model

Practical Tradeoffs & Production Thinking

7 questions

Multiple Choice

Model Versioning Strategy

Latency vs Accuracy Tradeoff

Feature Store Considerations

Model Retraining Frequency

Open-Ended

Trading Accuracy for Inference Speed

When Simpler Models Win

Detecting Model and Data Drift