# A machine-learning approach to Detect users' suspicious behaviour through the Facebook wall

**Authors:** Aimilia Panagiotou, Bogdan Ghita, Stavros Shiaeles, Keltoum Bendiab  
**arXiv:** [1910.14417](https://arxiv.org/abs/1910.14417)  
**Date:** 2019-10-31  
**Summary page:** https://www.summarizepaper.com/en/paper/1910.14417v1/

## Key points

- Social media platforms like Facebook have transformed the way people interact and communicate with each other.
- Increased reliance on social media means users are sharing more personal information than ever before, which can be used to gather intelligence through Open Source Intelligence (OSINT) techniques.
- Researchers propose a new method for detecting variations in overall Facebook user psychology using machine learning and OSINT.
- Their approach involves aggregating user sentiments and views over time using N-Games charts, which exhibit noticeable variations that can be validated through long-term collection.
- Security organizations can gain insights into user psychology by analyzing these charts and use the information to predict insider threats or prevent insider attacks.
- The authors argue that their approach could help identify suspicious behavior patterns among users that may indicate potential insider threats or attacks.
- This research highlights the importance of understanding how social media platforms like Facebook are changing the nature of social relationships and how this information can be used for both positive and negative purposes.

## Summary

The rise of social media platforms like Facebook has transformed the way people interact and communicate with each other. However, this increased reliance on social media also means that users are sharing more personal information than ever before, which can be used to gather intelligence through Open Source Intelligence (OSINT) techniques. In response to this, a team of researchers has proposed a new method for detecting variations in overall Facebook user psychology using machine learning and OSINT. Their approach involves aggregating user sentiments and views over time using N-Games charts, which exhibit noticeable variations that can be validated through long-term collection. By analyzing these charts, security organizations can gain insights into user psychology and use the information to predict insider threats or prevent insider attacks. In their paper titled "A machine-learning approach to Detect users' suspicious behaviour through the Facebook wall," Aimilia Panagiotou, Bogdan Ghita, Stavros Shiaeles, and Keltoum Bendiab outline how their method could be used by security organizations to understand and evaluate user behavior on Facebook. The authors argue that their approach could help identify suspicious behavior patterns among users that may indicate potential insider threats or attacks. Overall, this research highlights the importance of understanding how social media platforms like Facebook are changing the nature of social relationships and how this information can be used for both positive and negative purposes. By developing new methods for analyzing user behavior on these platforms, researchers hope to improve our ability to detect potential security threats and protect against them in the future.