Feat : Add naive_bayes

[MeKaustubh07]

🧠 Naive Bayes Classifier (R Implementation)

Overview

Naive Bayes is a probabilistic machine learning algorithm based on Bayes’ Theorem, assuming that all features are independent given the class.
It’s simple, fast, and works well for classification tasks like text analysis, spam detection, and document categorization.

---

Algorithm Concept

For a given input ( X = (x_1, x_2, ..., x_n) ), the probability of belonging to a class ( C_k ) is:

[
P(C_k | X) \propto P(C_k) \prod_{i=1}^{n} P(x_i | C_k)
]

The algorithm predicts the class with the highest posterior probability.

---

Variants Implemented

1. Gaussian Naive Bayes (GaussianNB)

   • For continuous data
   • Assumes features follow a normal distribution

2. Multinomial Naive Bayes (MultinomialNB)

   • For count-based features (e.g., word frequencies)
   • Commonly used in text classification

3. Bernoulli Naive Bayes (BernoulliNB)

   • For binary features (e.g., word presence/absence)
   • Ideal for spam detection and document filtering

---

Key Functions

# Train the model
fit(X, y)

# Predict class labels
predict(X)

# Predict class probabilities
predict_proba(X)

# Evaluate accuracy
score(X, y)

[Copilot]

Pull Request Overview

This PR adds a comprehensive implementation of the Naive Bayes classification algorithm to the machine learning directory. The implementation includes three variants of Naive Bayes classifiers optimized for different types of data.

• Implements Gaussian, Multinomial, and Bernoulli Naive Bayes variants using R6 classes
• Provides comprehensive examples demonstrating each variant with appropriate datasets
• Includes proper probability calculations with numerical stability considerations

[Copilot]

Pull Request Overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 1 comment.

[Copilot]

Pull Request Overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 2 comments.

[siriak]

Looks good, thanks!