Abstract
The number of software vulnerabilities discovered and publicly disclosed is increasing every year; however, only a small fraction of them is exploited in real-world attacks. With limitations on time and skilled resources, organizations often look at ways to identify threatened vulnerabilities for patch prioritization. In this paper, we present an exploit prediction model that predicts whether a vulnerability will be exploited. Our proposed model leverages data from a variety of online data sources (white-hat community, vulnerability researchers community, and darkweb/deepweb sites) with vulnerability mentions. Compared to the standard scoring system (CVSS base score), our model outperforms the baseline models with an F1 measure of 0.40 on the minority class (266% improvement over CVSS base score) and also achieves high True Positive Rate at low False Positive Rate (90%, 13%, respectively). The results demonstrate that the model is highly effective as an early predictor of exploits that could appear in the wild. We also present a qualitative and quantitative study regarding the increase in the likelihood of exploitation incurred when a vulnerability is mentioned in each of the data sources we examine.
| Original language | English (US) |
|---|---|
| Title of host publication | 2017 IEEE International Conference on Cyber Conflict U.S., CyCon U.S. 2017 - Proceedings |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 82-88 |
| Number of pages | 7 |
| Volume | 2017-December |
| ISBN (Electronic) | 9781538623794 |
| DOIs | |
| State | Published - Dec 5 2017 |
| Event | 2017 IEEE International Conference on Cyber Conflict U.S., CyCon U.S. 2017 - Washington, United States Duration: Nov 7 2017 → Nov 8 2017 |
Other
| Other | 2017 IEEE International Conference on Cyber Conflict U.S., CyCon U.S. 2017 |
|---|---|
| Country | United States |
| City | Washington |
| Period | 11/7/17 → 11/8/17 |
Fingerprint
Keywords
- adversarial machine learning
- darkweb analysis
- online vulnerability mentions
- vulnerability exploit prediction
ASJC Scopus subject areas
- Safety, Risk, Reliability and Quality
- Political Science and International Relations
- Computer Networks and Communications
- Law
Cite this
Proactive identification of exploits in the wild through vulnerability mentions online. / Almukaynizi, Mohammed; Nunes, Eric; Dharaiya, Krishna; Senguttuvan, Manoj; Shakarian, Jana; Shakarian, Paulo.
2017 IEEE International Conference on Cyber Conflict U.S., CyCon U.S. 2017 - Proceedings. Vol. 2017-December Institute of Electrical and Electronics Engineers Inc., 2017. p. 82-88.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Proactive identification of exploits in the wild through vulnerability mentions online
AU - Almukaynizi, Mohammed
AU - Nunes, Eric
AU - Dharaiya, Krishna
AU - Senguttuvan, Manoj
AU - Shakarian, Jana
AU - Shakarian, Paulo
PY - 2017/12/5
Y1 - 2017/12/5
N2 - The number of software vulnerabilities discovered and publicly disclosed is increasing every year; however, only a small fraction of them is exploited in real-world attacks. With limitations on time and skilled resources, organizations often look at ways to identify threatened vulnerabilities for patch prioritization. In this paper, we present an exploit prediction model that predicts whether a vulnerability will be exploited. Our proposed model leverages data from a variety of online data sources (white-hat community, vulnerability researchers community, and darkweb/deepweb sites) with vulnerability mentions. Compared to the standard scoring system (CVSS base score), our model outperforms the baseline models with an F1 measure of 0.40 on the minority class (266% improvement over CVSS base score) and also achieves high True Positive Rate at low False Positive Rate (90%, 13%, respectively). The results demonstrate that the model is highly effective as an early predictor of exploits that could appear in the wild. We also present a qualitative and quantitative study regarding the increase in the likelihood of exploitation incurred when a vulnerability is mentioned in each of the data sources we examine.
AB - The number of software vulnerabilities discovered and publicly disclosed is increasing every year; however, only a small fraction of them is exploited in real-world attacks. With limitations on time and skilled resources, organizations often look at ways to identify threatened vulnerabilities for patch prioritization. In this paper, we present an exploit prediction model that predicts whether a vulnerability will be exploited. Our proposed model leverages data from a variety of online data sources (white-hat community, vulnerability researchers community, and darkweb/deepweb sites) with vulnerability mentions. Compared to the standard scoring system (CVSS base score), our model outperforms the baseline models with an F1 measure of 0.40 on the minority class (266% improvement over CVSS base score) and also achieves high True Positive Rate at low False Positive Rate (90%, 13%, respectively). The results demonstrate that the model is highly effective as an early predictor of exploits that could appear in the wild. We also present a qualitative and quantitative study regarding the increase in the likelihood of exploitation incurred when a vulnerability is mentioned in each of the data sources we examine.
KW - adversarial machine learning
KW - darkweb analysis
KW - online vulnerability mentions
KW - vulnerability exploit prediction
UR - http://www.scopus.com/inward/record.url?scp=85045989630&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85045989630&partnerID=8YFLogxK
U2 - 10.1109/CYCONUS.2017.8167501
DO - 10.1109/CYCONUS.2017.8167501
M3 - Conference contribution
AN - SCOPUS:85045989630
VL - 2017-December
SP - 82
EP - 88
BT - 2017 IEEE International Conference on Cyber Conflict U.S., CyCon U.S. 2017 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
ER -