Skip to content

README.md

Latest commit

 

History

History
32 lines (21 loc) · 1.5 KB

README.md

File metadata and controls

32 lines (21 loc) · 1.5 KB
Raw

Email_Text_Classification

By: Matthew Walter, Eric Allen, Murugesan Somasundaram

Libraries: os, sys, re, math, glob, collections, pandas, numpy,

Summary: Implement and evaluate Naive Bayes, Perceptron, and Logistic Regression for text classification.

Requires 5-7 additional arguments:

python main.py <path_ham_train> <path_spam_train> <path_ham_test> <path_spam_test> <OPTIONAL: lambda> <OPTIONAL: iterations>

program: BAYES | MCAP | PERCEPTRON
path_ham_train: Directory Path
path_spam_train: Directory Path
path_ham_test: Directory Path
path_spam_test: Directory Path
lambda: Float Value (0.01 or 0.1 is recommended)
iterations: Integer Value

To Run

python main.py BAYES <path_ham_train> <path_spam_train> <path_ham_test> <path_spam_test>

python main.py MCAP <path_ham_train> <path_spam_train> <path_ham_test> <path_spam_test>

python main.py PERCEPTRON <path_ham_train> <path_spam_train> <path_ham_test> <path_spam_test>

Example Run

python main.py BAYES hw2datasets/dataset1/train/ham/ hw2datasets/dataset1/train/spam/ hw2datasets/dataset1/test/ham/ hw2datasets/dataset1/test/spam/

python main.py MCAP hw2datasets/dataset1/train/ham/ hw2datasets/dataset1/train/spam/ hw2datasets/dataset1/test/ham/ hw2datasets/dataset1/test/spam/ 0.01 100

python main.py PERCEPTRON hw2datasets/dataset1/train/ham/ hw2datasets/dataset1/train/spam/ hw2datasets/dataset1/test/ham/ hw2datasets/dataset1/test/spam/ 1000