Preverjanje moči gesel: primerjava rešitve na osnovi pametnih algoritmov in markovskih verig z obstoječimi spletnimi merilniki<!-- notionvc: 1d6644fe-17b6-4aef-8982-10b7890028ea -->

Despite the rise of alternative authentication methods, passwords remain the most widely used mechanism, making their strength evaluation critical. Due to the limitations of current password strength meters, this paper explores a more advanced, probabilistic approach. We present a web-based password strength evaluation tool based on Markov chains. The tool combines three models, each trained on a different dataset, and calculates the average score as the final strength estimate. The system was tested on a dataset of 10,000 passwords and benchmarked against nine popular online meters. Additionally, password cracking resistance was assessed using probabilistic context-free grammars. Results show that the proposed tool performs competitively, although it still lags slightly behind top-performing solutions. This confirms the practical value of statistical modelling in enhancing password security.