.png){kind=icon}
Return
Developing adaptive password strength algorithms using deep learning for enhanced cybersecurity
.png){kind=icon}
PRE
AI
DOI：10.47974/JDMSC-2502.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
research Abstract gthOftenTheexploitedsystembyuesbrute-forceconvolutional, andrecurrent,credential-stuffingand attacks,transformemodernr-basedprotectionneural architecturesto bettergraspthepatternsinthesyntaxandmeaningof methods are still generally weak and simple to guess passwords. Because they are constant paswordsToadapttonewpasswordpatternsandattackpaththemodel and ignore the context, traditional password strength meters-which rely on rule-based employs massive real-worldandmade-updatasetacharacter-levelembedding formulae or entropy calculations-are becoming poorer at assessing howsafe a password approachdynamcscoringsystems andcontinuous learning Theproposed is in the real world. The research offers a flexible, deep learning-based approach to assess modeloutperforms standardandbaseline classifiers bya wide marginintrals in password strength. The system uses convolutional, recurrent, and transformer-based neural accuracy (91 6%) precision (902%)recall(897%) andROCAUC(0956) architectures to better grasp the patterns in the syntax and meaning of passwords. To adapt Case tudies revealthat the model can identify false passwordpatterns and to new password patterns and attack paths, the model employs massive real-world and provide more accu ate power ev luations made-up datasets, a character-level embedding approach, dynamic scoring systems, and continuous learning. The proposed model outperforms standard and baseline classifiers by S bj t Cla ificatio 68M25 a wide margin in trials in accuracy (91.6%), precision (90.2%), recall (89.7%), and ROC-AUC K ywo d P d Str nth Esti ti C be rit De L i g (0.956). Case studies reveal that the model can identify false password patterns and provide Ad pti L i g Col i l N l N w k (CNN) more accurate power evaluations.
Keywords:
Password strength estimation
Cybersecurity
Deep learning
Adaptive learning
Convolutional neural networks (CNN)
Journal
J
IF:
1.1
Papers:
185
Citations:
0
Organization
View more organizations
Cited Papers
No cited papers available