基于贝叶斯知识追踪的网安人才能力智能化评估方法

张方娇; 赵建军; 刘心宇; 王晓蕾; 刘奇旭; 崔翔

引用本文：

张方娇,赵建军,刘心宇,王晓蕾,刘奇旭,崔翔.基于贝叶斯知识追踪的网安人才能力智能化评估方法[J].信息安全学报,2021,6(1):62-77 [点击复制]
ZHANG Fangjiao,ZHAO Jianjun,LIU Xinyu,WANG Xiaolei,LIU Qixu,CUI Xiang.Cybersecurity Talents Intelligent Evaluation Based on Bayesian Knowledge Tracing Model[J].Journal of Cyber Security,2021,6(1):62-77 [点击复制]

本文已被：浏览 7553次下载 5406次	码上扫一扫！
基于贝叶斯知识追踪的网安人才能力智能化评估方法
张方娇^1,2, 赵建军^1,2, 刘心宇^1,2, 王晓蕾^1,2, 刘奇旭^1,2, 崔翔³
0 字体:加大+\|默认\|缩小-
(1.中国科学院信息工程研究所北京中国 100093;2.中国科学院大学网络空间安全学院北京中国 100049;3.广州大学网络空间先进技术研究院广州中国 510006中国科学院信息工程研究所北京中国 100093)

摘要:

近年来，网络空间安全形势日益严峻，导致网络空间安全人才（以下简称网安人才）缺口巨大，国家加快网安人才评估的需求愈加强烈。针对当前网安人才能力评估精准度不足的问题，本文提出了一种改进的贝叶斯知识追踪CT-BKT （CybersecurityTalents Bayesian Knowledge Tracing）模型，通过网安人才能力评估时的个性智能化问答过程，该模型可对网安人才的知识状态进行追踪，从而实现对其能力的动态精准评估。为了验证CT-BKT模型的有效性，本文以Web安全为例，梳理了Web安全的知识技能体系并构建了相应题库，实现了一个面向Web安全领域的网安人才技能智能化评估系统CTIES （Cybersecurity Talents Intelligent Evaluation System）。通过对22名网安人员进行Web安全的能力评估，本文提出的CT-BKT知识追踪模型的对网安人才的知识掌握状态的预测准确率较高，CTIES系统能细致且直观地展现网安人才Web安全的知识掌握程度及相应专业技能水平，验证了本文所提出的网安人才能力评估方法的可行性和有效性。

关键词: 贝叶斯网络知识追踪模型网安人才智能化评估 Web安全

DOI：10.19363/J.cnki.cn10-1380/tn.2021.01.06

投稿时间：2020-09-30修订日期：2020-11-29

基金项目:中国科学院网络测评技术重点实验室和网络安全防护技术北京市重点实验室资助。

Cybersecurity Talents Intelligent Evaluation Based on Bayesian Knowledge Tracing Model

ZHANG Fangjiao^1,2, ZHAO Jianjun^1,2, LIU Xinyu^1,2, WANG Xiaolei^1,2, LIU Qixu^1,2, CUI Xiang³

(1.Depart Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093,China;2.School of Cyber Security, University of Chinese Academy of Sciences, Beijing 100049, China;3.Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou 510006, ChinaDepart Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093,China)

Abstract:

In recent years, the situation of cyberspace security is becoming more and more serious, which leads to a huge gap of cybersecurity talents. And our country has an increasingly strong demand for cybersecurity talents. To solve the accuracy in cybersecurity talents evaluation, this paper proposes an improved Bayesian knowledge tracking CT-BKT (Cybersecurity Talents Bayesian Knowledge Tracing) model. By tracking the knowledge status of cybersecurity talents, the individualized questions can be intelligently generated, so that dynamic and accurate evaluation of their capabilities can be achieved according to their answering to selected questions. In order to verify the effectiveness of CT-BKT model, here Web security is taken as an example. We sort out the knowledge and skills of Web security and construct the corresponding question bank, and finally implements a Cybersecurity Talents Intelligent Evaluation system (CTIES) in the Web security field. Through the evaluation of 22 cybersecurity talents, CT-BKT knowledge tracking model proposed in this paper has a high prediction accuracy. Besides, CTIES system can give the more detailed and direct display of cybersecurity talents’ knowledge mastery of Web security and corresponding professional skill level. The experiment verifies the feasibility and effectiveness of cybersecurity talents evaluation proposed in this paper.

Key words: bayesian network knowledge tracing model cybersecurity talents intelligent evaluation web security