Scholarlink

This research study proposes the analysis of several machine learning models for calculating password strength Among the models applied, decision trees, random forests, and AdaBoost achieved high predictive accuracy, reaching up to 84%. These models show promise in practical applications for classifying passwords into different strength levels. KNN, though less accurate, has the benefit of shorter train times and remains usable. In comparison, SVM as well as Naïve Bayes suffered from defects in classification, especially weak password cases, and could not train easily. This suggests the importance of machine learning approach in assessing password strength, such as balancing accuracy with computational efficiency and overall model performance. With the increasing cybercrime, it is critical that such research can be applied towards enhancing security and protecting sensitive information.

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

Implementing Machine Learning Based Strength Assessment Techniques to Enhance Password Security

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Tiyas Sarkar,
  • Manik Rakhra

摘要

This research study proposes the analysis of several machine learning models for calculating password strength Among the models applied, decision trees, random forests, and AdaBoost achieved high predictive accuracy, reaching up to 84%. These models show promise in practical applications for classifying passwords into different strength levels. KNN, though less accurate, has the benefit of shorter train times and remains usable. In comparison, SVM as well as Naïve Bayes suffered from defects in classification, especially weak password cases, and could not train easily. This suggests the importance of machine learning approach in assessing password strength, such as balancing accuracy with computational efficiency and overall model performance. With the increasing cybercrime, it is critical that such research can be applied towards enhancing security and protecting sensitive information.