Implementation of Naive Bayes from scratch using Python.
- This is a python implementation of Naïve Bayes Algorithm (m-estimate version) for a binary classification problem from scratch.
- The program will take two inputs: a training dataset and a test dataset.
- The value of m will be the number of possible values of a feature.
- All features will be categorical in the dataset.
- The last column of the datasets will be the class variable.
- The two-class values as positive and negative.
- The program will first load the training dataset, calculate all required probabilities, and then predict the class of each instance in the test set.
- The program will add a new column containing predicted classes to the test dataset and save it as predictions.csv.
- The program will also show the accuracy, sensitivity, and specificity on the console