Multiclass Password Strength Classification (MPSC 2024) Dataset for password cracking detection and prevention

We present a novel cutting-edge, large-scale multiclass dataset to improve the security of password protected systems cognition of suspicious password cracking attempts. The proposed newly generated dataset contains up-to-date samples and features available to the public to help reduce the effect of upcoming cyberattacks with machine learning methods. Specifically, 700,000 samples with more than 100 features are collected, processed through several stages including hashing, tokenization (NLP techniques), an others, and organized into three password classes: weak, moderate, and strong. For detailed info, Please refer to and cite our articles: Al-Haija, Q.A., Abu-Ghazaleh, R., Hafez, A., Mansour, S., Aljammal, Y. (2024). Password Security: Cracking Techniques and Countermeasures. Proceedings of Data Analytics and Management. ICDAM 2024. Lecture Notes in Networks and Systems, Springer, 2024. Al-Haija, Q.A., Abu-Ghazaleh, R., Hafez, A., Mansour, S., Aljammal, Y. (2024). PasswordProtectorPro: A Password Cracking Detection and Prevention Tool for Mission Critical Systems. 8th IET Smart Cities Symposium (SCS 2024), Hybrid Conference, Bahrain, 2024.

本研究提出一款前沿新颖的大规模多分类数据集，旨在提升密码保护系统识别可疑密码破解尝试的认知水平与安全防护能力。本数据集为全新构建，包含公开可用的最新样本与特征集，可借助机器学习方法缓解未来网络攻击带来的影响。具体而言，本数据集共收集70万个样本、超100项特征，历经哈希（hashing）、分词（Tokenization，自然语言处理（NLP）技术）等多阶段处理流程，并将样本划分为弱密码、中等强度密码、强密码三类。如需获取详细信息，请引用并参考以下研究成果：Al-Haija, Q.A., Abu-Ghazaleh, R., Hafez, A., Mansour, S., Aljammal, Y. (2024). "Password Security: Cracking Techniques and Countermeasures". Proceedings of Data Analytics and Management. ICDAM 2024. Lecture Notes in Networks and Systems, Springer, 2024. Al-Haija, Q.A., Abu-Ghazaleh, R., Hafez, A., Mansour, S., Aljammal, Y. (2024). "PasswordProtectorPro: A Password Cracking Detection and Prevention Tool for Mission Critical Systems". 8th IET Smart Cities Symposium (SCS 2024), Hybrid Conference, Bahrain, 2024.

搜集汇总

背景与挑战

背景概述

该数据集是一个大规模多类密码强度分类数据集，包含70万个样本和超过100个特征，用于密码破解检测和预防。样本经过哈希和标记化等处理，分为弱、中等和强三个类别，旨在通过机器学习方法增强密码系统的安全性，以应对网络攻击威胁。