Scribd
Upload
119 views

SMS Spam Detection Presentation

The SMS Spam Detection Project explores the use of machine learning to identify and filter spam messages from inboxes. Various models, including Naive Bayes, Logistic Regression, and Random Forest, were evaluated, with Logistic Regression achieving the highest accuracy of 95%. The project highlights the importance of feature engineering and suggests future work on AI-powered chatbots for enhanced spam filtering.

Uploaded by

ravitejabogireddy12
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
119 views

SMS Spam Detection Presentation

The SMS Spam Detection Project explores the use of machine learning to identify and filter spam messages from inboxes. Various models, including Naive Bayes, Logistic Regression, and Random Forest, were evaluated, with Logistic Regression achieving the highest accuracy of 95%. The project highlights the importance of feature engineering and suggests future work on AI-powered chatbots for enhanced spam filtering.

Uploaded by

ravitejabogireddy12
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

SMS Spam Detection Project

A Fun and Insightful Approach


Ravi Teja
Problem Statement
• Spam messages flood our inboxes daily. But
can we teach a machine to detect and filter
them?

• Let's explore!
Dataset & Preprocessing
• • Collected SMS spam dataset
• • Removed stopwords and punctuations
• • Converted text into numerical features
• • Used TF-IDF for feature extraction
Machine Learning Models
• • Naive Bayes (Fast and simple)
• • Logistic Regression (Balanced approach)
• • Random Forest (More accurate but heavier)
• • Deep Learning (Neural Networks for
advanced filtering)
Feature Engineering
• • Word frequency analysis
• • N-grams for phrase detection
• • Character length of messages
• • Presence of spammy words like 'win', 'prize',
'free', etc.
Model Evaluation
• • Accuracy: Measures overall correctness
• • Precision: How many detected spam
messages were actually spam?
• • Recall: How well did we catch all spam
messages?
• • F1-Score: The balance between Precision
and Recall
Results & Insights
• • Logistic Regression performed best with 95%
accuracy!
• • Random Forest had better recall but took
longer to train.
• • Word-based features played a crucial role in
spam detection.
Conclusion & Future Work
• • Spam detection can save time and protect
users!
• • Future work: Implementing AI-powered
chatbots for spam filtering.
• • Remember: If a message says 'You won a
lottery!'... you probably didn't.

You might also like