Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Recurso digital |
| Sprache: | |
| Veröffentlicht: |
Zenodo
2022
|
| Online-Zugang: | https://doi.org/10.5281/zenodo.15096230 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866902013583818752 |
|---|---|
| author | Raghavender Maddali |
| author_facet | Raghavender Maddali |
| contents | <p>The fast evolution of digital infrastructures and growing sophistication of cyber-attacks, conventional security functionalities always fail to ensure real-time protection. The current study presents a threat detection framework based on machine learning to improve data security by identifying anomalies, behavior analysis, and predictive modeling. Proactive detection and prevention of cyber-attacks are facilitated by the new framework, minimizing probable threats and providing strong security functionalities. It applies supervised and unsupervised learning methods for anomaly detection of normal behavior to have early threat recognition and response. Deep learning models, added in the system, increase accuracy for recognizing complex patterns of attacks. The design is highly flexible in all industries and provides data security as well as regulation compliance. It is illustrated through performance testing that it has high detection capability with low false positives, and therefore, it is an effective solution for real-time security for sensitive information. The research also points out the need for AI-based cybersecurity methods in evading emerging cyber threats.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_15096230 |
| institution | Zenodo |
| language | |
| publishDate | 2022 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Enhancing Data Security with Machine Learning-Driven Threat Detection Raghavender Maddali <p>The fast evolution of digital infrastructures and growing sophistication of cyber-attacks, conventional security functionalities always fail to ensure real-time protection. The current study presents a threat detection framework based on machine learning to improve data security by identifying anomalies, behavior analysis, and predictive modeling. Proactive detection and prevention of cyber-attacks are facilitated by the new framework, minimizing probable threats and providing strong security functionalities. It applies supervised and unsupervised learning methods for anomaly detection of normal behavior to have early threat recognition and response. Deep learning models, added in the system, increase accuracy for recognizing complex patterns of attacks. The design is highly flexible in all industries and provides data security as well as regulation compliance. It is illustrated through performance testing that it has high detection capability with low false positives, and therefore, it is an effective solution for real-time security for sensitive information. The research also points out the need for AI-based cybersecurity methods in evading emerging cyber threats.</p> |
| title | Enhancing Data Security with Machine Learning-Driven Threat Detection |
| url | https://doi.org/10.5281/zenodo.15096230 |