Supervised learning is only used for classification tasks.

Supervised learning uses labeled data for training, while unsupervised learning uses unlabeled data to find patterns.

Supervised learning is faster than unsupervised learning.

Unsupervised learning requires more data than supervised learning.

Logistic regression is used for binary classification problems.

Clustering problems

Regression analysis for continuous outcomes

Time series forecasting

KNN classifies data points by random selection.

The KNN algorithm classifies data points based on the majority class of their 'k' nearest neighbors.

KNN uses a single farthest neighbor to classify points.

KNN assigns classes based on the average distance to all points.

Support Vector Method

Support Vector Model

Supervised Vector Machine

Support Vector Machine

Random forests have lower accuracy than decision trees in all cases.

Decision trees are always more accurate than random forests.

Decision trees and random forests have the same level of accuracy.

Random forests are generally more accurate than decision trees.

y = mx + b

y = m + bx

y = ax^2 + b

y = mx^2 + c

It averages the coordinates of all points to determine proximity.

The KNN algorithm determines the nearest neighbors by calculating distances and selecting the closest 'k' points.

It uses a decision tree to find neighbors.

It randomly selects points without considering distance.

Requires large datasets for accuracy

High complexity in model interpretation

Inability to handle categorical variables

Advantages of using logistic regression include simplicity, interpretability, efficiency, probability outputs, and suitability for binary outcomes.

When computational efficiency is the top priority

When the data is not linearly separable

When working with small datasets and noisy data

When dealing with high-dimensional data, limited data, need for clear margins, or larger datasets.

Overfitting in decision trees is when the model captures noise and outliers, leading to poor generalization on new data.

Overfitting occurs when the model is too simple and fails to capture the underlying patterns.

Overfitting is when the model performs well on training data but poorly on validation data due to lack of complexity.

Overfitting happens when the decision tree is pruned excessively, leading to underfitting.

Random forest reduces overfitting and increases accuracy by combining multiple decision trees.

Random forest uses a single decision tree for predictions.

Random forest does not require any data preprocessing.

Random forest only works with categorical data.

The cost function quantifies the error between predicted and actual values, guiding the optimization of model parameters.

The cost function is used to calculate the correlation coefficient.

The cost function determines the model's complexity.

The cost function predicts future values directly.

The margin is the average distance of all points from the decision boundary.

The margin is the total number of support vectors used in classification.

The margin in SVM classification is the distance between the closest points of different classes to the decision boundary, which SVM aims to maximize.

The margin is the area where no data points exist in the feature space.

learning_rate

dropout_rate

n_estimators, max_depth, min_samples_split, max_features

batch_size