CyberSafeSolutions

Spam Detection Model

Overview

This project implements a spam detection system using the Multinomial Naive Bayes algorithm. It preprocesses text data, balances the dataset, and uses TF-IDF (Term Frequency-Inverse Document Frequency) for feature extraction. The model is fine-tuned using GridSearchCV and evaluated using metrics such as accuracy, precision, recall, and F1-score. A real-time prediction function is also provided to classify messages as spam or legitimate.

---

Features

• Text Preprocessing: Converts text to lowercase, removes punctuation and numbers, and applies stemming and stopword removal.
• Dataset Balancing: Upsamples the minority class (spam) to balance the dataset.
• Feature Extraction: Uses TF-IDF to convert text data into numerical features.
• Model Tuning: Utilizes GridSearchCV to find the best hyperparameters for the Multinomial Naive Bayes model.
• Real-Time Prediction: Classifies new messages as spam or legitimate.

---

Requirements

• Python libraries: pandas, scikit-learn, nltk
• NLTK stopwords dataset (download using nltk.download('stopwords')).

---

Dataset

The dataset (spam.csv) contains two columns:

• label: Indicates whether a message is spam (spam) or legitimate (ham).
• message: The text content of the message.

---

Performance Metrics

The best model achieved the following results:

• Accuracy: 97.62%
• Precision (Spam): 95.82%
• Recall (Spam): 99.47%
• F1-Score (Spam): 97.61%

Confusion Matrix

	Predicted Ham	Predicted Spam
Actual Ham	944	41
Actual Spam	5	940

---

Real-Time Prediction

A function predict_spam is provided to classify new messages. For example:

message = "Congratulations! You've won a $1000 Walmart gift card. Click here to claim now."
result = predict_spam(message)  # Output: Spam

---