Explainable Artificial Intelligence-Based Model Drift Detection Applicable to Unsupervised Environments

作     者：Yongsoo Lee Yeeun Lee Eungyu Lee Taejin Lee 

作者机构：Department of Information SecurityHoseo UniversityAsan31499Korea 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2023年第76卷第8期

页      面：1701-1719页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 0839[工学-网络空间安全] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the Institute of Information and Communications Technology Planning and Evaluation(IITP)grant funded by the Korean government(MSIT)(No.2022-0-00089,Development of clustering and analysis technology to identify cyber attack groups based on life cycle) the Institute of Civil Military Technology Cooperation funded by the Defense Acquisition Program Administration and Ministry of Trade,Industry and Energy of Korean government under Grant No.21-CM-EC-07 

主　　题：Cybersecurity machine learning(ML) model life-cycle management drift detection unsupervised environments shapley additive explanations(SHAP) explainability 

摘      要：Cybersecurity increasingly relies on machine learning(ML)models to respond to and detect ***,the rapidly changing data environment makes model life-cycle management after deployment ***-time detection of drift signals from various threats is fundamental for effectively managing deployed ***,detecting drift in unsupervised environments can be *** study introduces a novel approach leveraging Shapley additive explanations(SHAP),a widely recognized explainability technique in ML,to address drift detection in unsupervised *** proposed method incorporates a range of plots and statistical techniques to enhance drift detection reliability and introduces a drift suspicion metric that considers the explanatory aspects absent in the current *** validate the effectiveness of the proposed approach in a real-world scenario,we applied it to an environment designed to detect domain generation algorithms(DGAs).The dataset was obtained from various types of DGAs provided by *** on this dataset composition,we sought to validate the proposed SHAP-based approach through drift scenarios that occur when a previously deployed model detects new data types in an environment that detects real-world *** results revealed that more than 90%of the drift data exceeded the threshold,demonstrating the high reliability of the approach to detect drift in an unsupervised *** proposed method distinguishes itself fromexisting approaches by employing explainable artificial intelligence(XAI)-based detection,which is not limited by model or system environment *** conclusion,this paper proposes a novel approach to detect drift in unsupervised ML settings for *** proposed method employs SHAP-based XAI and a drift suspicion metric to improve drift detection *** is versatile and suitable for various realtime data analysis contexts beyond DGA detection *** study significan