This device is not compatible.

You will learn to:

Implement the Naïve-Bayes algorithm.

Compute the required probabilities using pandas.

Save and retrieve the probabilities using Python dictionaries.

Perform model evaluation using Scikit-learn.

Skills

Machine Learning

Data Science

Prerequisites

Good programming skills in Python.

Good understanding of Machine Learning theory.

Proficient in probability and statistics.

Technologies

NumPy

Python

Pandas

seaborn

Scikit-learn

Project Description

The Naïve-Bayes algorithm is known for its simplicity and efficiency in probabilistic classification. It uses the Bayes' rule to compute a target class's conditional probability of occurrence, given a set of input features. It is based on the naive assumption that classes of all features are mutually independent.

In this project, we'll implement the Naïve-Bayes classifier from scratch in Python, without using any external libraries. We will load the "US Census Dataset" and preprocess it. We’ll use the Scikit-learn library to evaluate the classifier. By implementing this classifier from scratch, we will gain a deeper understanding of its inner workings.

Project Tasks

1

Getting Started

Task 0: Introduction

Task 1: Import the Libraries

Task 2: Load the Dataset

Task 3: Preprocess the Data

2

Implement the Naïve-Bayes Classifier

Task 4: The Initialization Method

Task 5: Outlier Handler

Task 6: Convert Numeric Features to Categorical

Task 7: Prepare Data

Task 8: The Train Function

Task 9: The Predict Function

3

Use the Model

Task 10: Model Creation, Training, and Prediction

Task 11: The Confusion Matrix

Task 12: Model Evaluation

Congratulations