Smart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligence
Abstract Cyber threats have become a global concern, and attackers use more efficient, hidden, and destructive techniques. Traditional signature- and rule-based systems struggle to encounter these latest threats, exposing organizations facing severe risks to their financial well-being and national s...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-05-01
|
| Series: | Discover Sustainability |
| Online Access: | https://doi.org/10.1007/s43621-025-01280-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850271043409674240 |
|---|---|
| author | Saif Jasim Almheiri Asghar Ali Shah Sagheer Abbas Munir Ahmad Muhammad Adnan Khan |
| author_facet | Saif Jasim Almheiri Asghar Ali Shah Sagheer Abbas Munir Ahmad Muhammad Adnan Khan |
| author_sort | Saif Jasim Almheiri |
| collection | DOAJ |
| description | Abstract Cyber threats have become a global concern, and attackers use more efficient, hidden, and destructive techniques. Traditional signature- and rule-based systems struggle to encounter these latest threats, exposing organizations facing severe risks to their financial well-being and national security risks. Although Artificial Intelligence (AI) provides techniques for detecting threats in real-time, its black-box nature introduces challenges related to trust, regulatory compliance, and interpretability, particularly in high-risk environments where decision transparency is critical. The black-box nature of AI systems in critical situations demands complete transparency and ease of interpretation for people to trust automated threat detection methods. In this research, the proposed Smart Sustainable Cybersecurity (SSC) model incorporates XAI to improve interpretability and assist in validating and improving threat classification models to reduce the number of false positives and make them more resistant to adversarial attacks. The proposed SSC-based XAI model had a high accuracy level of 99.65% and a low miss-rate of 0.35% compared to other approaches while offering greater transparency that may support resilience against adversarial behavior. While these results are promising, they are based on a controlled benchmark dataset and require further validation across diverse, real-world environments. This improves the detection of cyber threats and enhances the ability for intelligent, adaptive, and anticipative cyber defenses by minimizing false positives while advancing interpretability. |
| format | Article |
| id | doaj-art-6c0975bf13694335b9fa7b86aa184ffd |
| institution | OA Journals |
| issn | 2662-9984 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Springer |
| record_format | Article |
| series | Discover Sustainability |
| spelling | doaj-art-6c0975bf13694335b9fa7b86aa184ffd2025-08-20T01:52:22ZengSpringerDiscover Sustainability2662-99842025-05-016112010.1007/s43621-025-01280-zSmart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligenceSaif Jasim Almheiri0Asghar Ali Shah1Sagheer Abbas2Munir Ahmad3Muhammad Adnan Khan4School of Computing, Skyline University CollegeDepartment of Computer Science, Kateb UniversityDepartment of Computer Science, Prince Mohammad Bin Fahd UniversityUniversity College, Korea UniversitySchool of Computing, Skyline University CollegeAbstract Cyber threats have become a global concern, and attackers use more efficient, hidden, and destructive techniques. Traditional signature- and rule-based systems struggle to encounter these latest threats, exposing organizations facing severe risks to their financial well-being and national security risks. Although Artificial Intelligence (AI) provides techniques for detecting threats in real-time, its black-box nature introduces challenges related to trust, regulatory compliance, and interpretability, particularly in high-risk environments where decision transparency is critical. The black-box nature of AI systems in critical situations demands complete transparency and ease of interpretation for people to trust automated threat detection methods. In this research, the proposed Smart Sustainable Cybersecurity (SSC) model incorporates XAI to improve interpretability and assist in validating and improving threat classification models to reduce the number of false positives and make them more resistant to adversarial attacks. The proposed SSC-based XAI model had a high accuracy level of 99.65% and a low miss-rate of 0.35% compared to other approaches while offering greater transparency that may support resilience against adversarial behavior. While these results are promising, they are based on a controlled benchmark dataset and require further validation across diverse, real-world environments. This improves the detection of cyber threats and enhances the ability for intelligent, adaptive, and anticipative cyber defenses by minimizing false positives while advancing interpretability.https://doi.org/10.1007/s43621-025-01280-z |
| spellingShingle | Saif Jasim Almheiri Asghar Ali Shah Sagheer Abbas Munir Ahmad Muhammad Adnan Khan Smart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligence Discover Sustainability |
| title | Smart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligence |
| title_full | Smart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligence |
| title_fullStr | Smart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligence |
| title_full_unstemmed | Smart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligence |
| title_short | Smart sustainable cyber security: modelling an interpretable and transparent threat detection with explainable artificial intelligence |
| title_sort | smart sustainable cyber security modelling an interpretable and transparent threat detection with explainable artificial intelligence |
| url | https://doi.org/10.1007/s43621-025-01280-z |
| work_keys_str_mv | AT saifjasimalmheiri smartsustainablecybersecuritymodellinganinterpretableandtransparentthreatdetectionwithexplainableartificialintelligence AT asgharalishah smartsustainablecybersecuritymodellinganinterpretableandtransparentthreatdetectionwithexplainableartificialintelligence AT sagheerabbas smartsustainablecybersecuritymodellinganinterpretableandtransparentthreatdetectionwithexplainableartificialintelligence AT munirahmad smartsustainablecybersecuritymodellinganinterpretableandtransparentthreatdetectionwithexplainableartificialintelligence AT muhammadadnankhan smartsustainablecybersecuritymodellinganinterpretableandtransparentthreatdetectionwithexplainableartificialintelligence |