Scholarlink

Extended Detection and Response (XDR) is a platform that combines data from multiple sources to detect and respond to different security threats. They are designed to facilitate threat detection with improved visibility and reduced response times. False positives in Extended Detection and Response (XDR) systems are a fundamental challenge for cybersecurity frameworks, principally affecting operational efficiency, generating alert fatigue, and wasting limited resources. This study intends to conduct a comparative analysis of traditional rule-based detection mechanisms and evaluate how advanced techniques, like Machine Learning and Adaptive AI, can leveraged to reduce false positives with a simultaneous improvement in the detection accuracy. The study evaluates traditional rule-based and AI-driven detection techniques based on key performance metrics, including detection accuracy, false positive rate, operational efficiency, scalability, and ease of integration into existing security frameworks. The results demonstrate that AI-driven models, particularly those leveraging machine learning and adaptive AI, reduce false positives by up to 60%, improve analyst workload efficiency by 40%, and accelerate incident response times by approximately 50% compared to traditional methods. A comparative analysis approach was employed, using secondary data sources including industry reports, case studies to assess the effectiveness of traditional and advanced XDR techniques. This integration of AI in XDR systems holds some issues, such as difficulties in integration, data dependency, and model interpretability. This paper proposes a hybrid rule-based detection model to reduce these shortcomings that complement human cognition with partial automation, ensuring scalability, operational efficiency, and adherence to compliance. The research suggests the requirement for seamless and more intelligent XDR solutions and provides concrete recommendations to optimize XDR frameworks in a way that enhances the ability to fend off sophisticated cyber threats.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Reducing False Positives in XDR Systems: A Comparative Analysis of Traditional and Advanced Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Omkar Murkute,
  • T. N. Nisha

摘要

Extended Detection and Response (XDR) is a platform that combines data from multiple sources to detect and respond to different security threats. They are designed to facilitate threat detection with improved visibility and reduced response times. False positives in Extended Detection and Response (XDR) systems are a fundamental challenge for cybersecurity frameworks, principally affecting operational efficiency, generating alert fatigue, and wasting limited resources. This study intends to conduct a comparative analysis of traditional rule-based detection mechanisms and evaluate how advanced techniques, like Machine Learning and Adaptive AI, can leveraged to reduce false positives with a simultaneous improvement in the detection accuracy. The study evaluates traditional rule-based and AI-driven detection techniques based on key performance metrics, including detection accuracy, false positive rate, operational efficiency, scalability, and ease of integration into existing security frameworks. The results demonstrate that AI-driven models, particularly those leveraging machine learning and adaptive AI, reduce false positives by up to 60%, improve analyst workload efficiency by 40%, and accelerate incident response times by approximately 50% compared to traditional methods. A comparative analysis approach was employed, using secondary data sources including industry reports, case studies to assess the effectiveness of traditional and advanced XDR techniques. This integration of AI in XDR systems holds some issues, such as difficulties in integration, data dependency, and model interpretability. This paper proposes a hybrid rule-based detection model to reduce these shortcomings that complement human cognition with partial automation, ensuring scalability, operational efficiency, and adherence to compliance. The research suggests the requirement for seamless and more intelligent XDR solutions and provides concrete recommendations to optimize XDR frameworks in a way that enhances the ability to fend off sophisticated cyber threats.