Password Strength Analyser and Security Recommendation System with Breach Intelligence and Encrypted Vault
Most digital accounts are still protected by passwords, and a large share of security breaches can be traced back to passwords that were weak, reused, or already leaked. Typical password meters only check shallow composition rules, such as a minimum length or a mix of character types, and these are easy to satisfy without ending up with a truly strong password. In this paper we present a hybrid Password Strength Analyser and Security Recommendation System. It combines Shannon-entropy estimation, structural pattern detection (keyboard walks, sequences, repeated characters and date-like substrings) and dictionary/common-password matching to give a realistic strength score from 0 to 100. The system also includes a privacy-preserving Breach Intelligence module built on the k-anonymity protocol, so only a truncated hash prefix of a password is ever sent outside the machine, and an AES-256-GCM encrypted credential vault that uses PBKDF2-HMAC-SHA256 key derivation for safe local storage. It is written in Java with no external dependencies and can be used through a console application or a browser-based web application running on the JDK's built-in HTTP server. Tests on a representative set of passwords show that the hybrid scoring approach can tell apart passwords that follow the composition rules but are still weak from passwords that are genuinely high in entropy, and that it does better than simple rule-based checks.
Authors
- Abhinav Soman*, Krushnraj Rajput, Rohan Gavit, Tushar Chinche, Jitendra Patil
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-15
- DOI
- https://doi.org/10.5281/zenodo.24251984
- Primary Topic
- User Authentication and Security Systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00