Gaussian Naive Bayes Classifier for Breast Cancer Detection

Introduction

This project implements a Gaussian Naive Bayes classifier to detect breast cancer based on the Breast Cancer Wisconsin dataset. The classifier is built using both scikit-learn's GaussianNB model and a custom Gaussian Naive Bayes implementation.

The Naive Bayes algorithm is a probabilistic classifier based on Bayes' Theorem, assuming that the features are conditionally independent given the class (i.e., the naive assumption). Gaussian Naive Bayes is a special case of Naive Bayes, which assumes that the continuous features follow a Gaussian (normal) distribution.

---

Dataset

The dataset used in this project is the Breast Cancer Wisconsin dataset, which is available in scikit-learn. It contains 569 instances and 30 features, where each instance is a patient's breast cancer diagnostic measurements. The target variable is binary, representing:

• 0: Malignant (cancerous)
• 1: Benign (non-cancerous)

Features:

• 30 real-valued input features, such as mean radius, mean texture, etc., calculated from a digitized image of a fine needle aspirate (FNA) of a breast mass.

---

Theory: Gaussian Naive Bayes

Naive Bayes is based on Bayes' Theorem, which calculates the posterior probability of a class C given a set of features X = {x1, x2, ..., xn}.

Bayes' Theorem:

$$ P(C | X) = \frac{P(X | C) \cdot P(C)}{P(X)} $$

Where:

• ( P(C | X) ): Posterior probability of class C given the feature vector X
• ( P(X | C) ): Likelihood of feature vector X given class C
• ( P(C) ): Prior probability of class C
• ( P(X) ): Evidence (total probability of X)

In Gaussian Naive Bayes, we assume that the likelihood ( P(X | C) ) is a Gaussian (normal) distribution, parameterized by the mean and variance of each feature in the training data for each class.

Gaussian Probability Density Function (PDF):

For a given feature x with mean μ and variance σ², the probability density function is given by:

$$ P(x | C) = \frac{1}{\sqrt{2\pi\sigma^2}} \cdot \exp\left(-\frac{(x - \mu)^2}{2\sigma^2}\right) $$

The Gaussian Naive Bayes algorithm combines the likelihoods of each feature for both classes and multiplies them with the class prior to predict the class with the highest posterior probability.

---

Project Workflow

1. Data Preprocessing

• Load the Breast Cancer dataset using scikit-learn's load_breast_cancer() function.
• Create a Pandas DataFrame containing the features and the target.
• Split the data into training (80%) and testing (20%) sets using train_test_split.

2. Custom Gaussian Naive Bayes Implementation

The custom implementation of Gaussian Naive Bayes performs the following steps:

1. Calculate mean and variance for each feature within each class on the training data.
2. Calculate prior probabilities based on the frequency of each class in the training set.
3. Predict class for new data by computing the posterior probability using Bayes' Theorem and Gaussian PDF.
4. Evaluate the accuracy of the custom implementation by comparing predictions to true labels.

3. Model Evaluation

• Accuracy is calculated as the proportion of correct predictions out of the total number of predictions.

4. scikit-learn Gaussian Naive Bayes

• Train the model using GaussianNB() from scikit-learn.
• Evaluate the model's accuracy on the test data.