# Modeling Suspicious Email Detection using Enhanced Feature Selection

**Authors:** Sarwat Nizamani, Nasrullah Memon, Uffe Kock Wiil, Panagiotis Karampelas  
**arXiv:** [1312.1971](https://arxiv.org/abs/1312.1971)  
**Date:** 2013-12-06  
**Summary page:** https://www.summarizepaper.com/en/paper/1312.1971v1/

## Key points

- The paper presents a model for detecting suspicious emails using enhanced feature selection.
- Feature selection strategies are proposed to be used in combination with classification techniques to identify terrorist emails.
- Machine learning algorithms such as decision tree (ID3), logistic regression, Naïve Bayes (NB), and Support Vector Machine (SVM) are evaluated in the model.
- The authors argue that implementing appropriate feature selection mechanisms can further improve the accuracy of email detection.
- A specific feature selection scheme is identified that enhances the performance of existing algorithms.
- Selecting relevant features is highlighted as important for improving the accuracy of email detection models.
- The proposed model aims to enhance the identification of suspicious emails and contribute to counterterrorism efforts.
- The research provides valuable insights into improving email detection systems for identifying potential threats and enhancing security measures.

## Summary

The paper "Modeling Suspicious Email Detection using Enhanced Feature Selection" presents a model for detecting suspicious emails by incorporating enhanced feature selection. The authors propose the use of feature selection strategies in combination with classification techniques to identify terrorist emails. The model focuses on evaluating machine learning algorithms such as decision tree (ID3), logistic regression, Naïve Bayes (NB), and Support Vector Machine (SVM) to detect emails containing suspicious content. While various algorithms have achieved good accuracy in this task, the authors argue that the results can be further improved by implementing appropriate feature selection mechanisms. They have identified a specific feature selection scheme that enhances the performance of existing algorithms. The paper highlights the importance of selecting relevant features to improve the accuracy of email detection models. By utilizing an effective feature selection strategy, the proposed model aims to enhance the identification of suspicious emails and contribute to counterterrorism efforts. Overall, this research provides valuable insights into improving email detection systems for identifying potential threats and enhancing security measures. The findings offer practical implications for developing more robust and accurate models for detecting suspicious content in emails.