QueryEd.com | Machine Learning Learning Hub

Review & Clarifications

Your understanding is a good start. I'll correct a few important points while keeping your explanations simple and intuitive.

1. Linear Regression

Linear regression is a machine learning model used to predict a numerical value by learning the relationship between input variables and an output variable.

• Purpose: Predict a numerical value.
• Example: Predict an exam score based on the number of hours studied.
• How: The model learns a best-fit line from training data. It uses this line to estimate the output for new input data.

2. Logistic Regression

Logistic regression is a machine learning model used to predict the probability of an outcome, usually when choosing between two classes, such as pass or fail.

• Purpose: Classify data into two classes.
• Example: Predict whether an email is spam or not spam.
• How: The model uses a sigmoid function to convert its output into a probability between 0 and 1. The sigmoid function produces an S-shaped curve. A threshold, often 0.5, can be used to select the predicted class.

3. Decision Tree

A decision tree is a machine learning model that makes predictions by following a series of questions or conditions.

• Purpose: Make predictions using a sequence of decisions.
• Example: Predict whether a customer will buy a product based on their age, income and other factors.
• How: Start at the root node and follow the branches according to the conditions. When you reach a leaf node, you get the final prediction.

A decision tree can predict categories or numerical values.

4. K-Nearest Neighbors (KNN)

K-Nearest Neighbours (KNN) is a machine learning algorithm that predicts the class or value of a new data point by looking at the nearest examples in the training data.

• Purpose: Classify data or predict numerical values using nearby examples.
• Example: Classify a flower by comparing its measurements with those of similar flowers.
• How: The model finds the K nearest data points. For classification, it often selects the class that appears most frequently among those neighbours.

K represents the number of nearest neighbours considered.

5. Random Forest

Random Forest is a machine learning model that combines multiple decision trees to make a prediction.

• Purpose: Make predictions by combining the results of many decision trees.
• Example: Predict whether an email is spam by combining the predictions of multiple trees.
• How: Each tree is trained using randomly sampled training data and considers randomly selected features when making splits.
• Classification: The trees vote for a class, and the model combines their predictions.
• Regression: The model usually averages the numerical predictions from its trees.

Remember the Differences

Model	Main Purpose
Linear Regression	Predict a numerical value
Logistic Regression	Predict a probability for classification
Decision Tree	Make predictions through decision rules
KNN	Predict using nearby examples
Random Forest	Combine many decision trees

💡

One important point: These models learn patterns from training data. They don't simply memorise a formula that works for every dataset.

Knowledge Check Quiz

10 questions covering all concepts above.

Mastering Machine Learning Essentials