Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2025
|
| Online Access: | https://doi.org/10.5281/zenodo.18152280 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <h2>Abstract</h2> <div>Email spam continues to be a significant challenge due to the increasing volume of unsolicited and malicious messages. This paper presents a comparative analysis of classical machine learning algorithms for email spam detection using the UCI Spambase dataset. Several widely used classifiers, including Naïve Bayes, Support Vector Machine, Logistic Regression, Decision Tree, and Random Forest, are evaluated under identical experimental conditions. Performance is measured by precision, recall, and F1-score ,accuracy. Experimental results demonstrate that the Random Forest classifier outperforms other models, achieving an accuracy of 94.57% with a high precision and balanced recall. In addition to performance evaluation, model interpretability is enhanced using SHAP (SHapley Additive exPlanations) to analyse feature contributions influencing spam classification decisions. The findings indicate that classical machine learning models, when combined with explainability techniques, can provide reliable and interpretable solutions for email spam filtering.</div> <h2><span></span>Keywords</h2> <div>Email Spam Detection, Machine Learning, Random Forest, SHAP, Text Classification.</div>