Cleartext Password Detection via Predictive Text Classifier
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting cleartext passwords are not effective, as they rely on known password lists and fail to identify unique or complex passwords, leaving security breaches unaddressed.
Innovation Solution
A predictive text classifier is trained on a corpus of text to identify password anomalies by analyzing the probability of character sequences, generating alerts for low-probability sequences that may indicate cleartext passwords, regardless of their similarity to known passwords.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If known password lists are used for detection, then simple passwords can be identified, but unique or complex passwords are missed
Solution Approach 1:
The patent changes the detection parameter from static password lists to dynamic probability thresholds. Instead of matching against fixed known passwords, the system calculates the probability of character sequences occurring in normal text and compares against a threshold, allowing detection of any password type regardless of whether it appears in pre-existing lists.
Solution Approach 2:
The patent replaces the mechanical approach of list-matching with a statistical probability model. Rather than mechanically comparing passwords against a database of known passwords, the system uses probabilistic language models to assess the likelihood of character sequences, substituting a more adaptable statistical mechanism.
2Reliability
If traditional detection methods are used, then existing security measures can be maintained, but labor intensive processes are required
Solution Approach 1:
The system performs self-service detection by automatically analyzing text files and source code without requiring manual inspection. The probabilistic model independently evaluates character sequences and generates detection results autonomously, eliminating the need for labor-intensive manual review while maintaining reliable security detection.
Solution Approach 2:
The patent substitutes manual mechanical inspection with automated probabilistic analysis. The system uses computational probability models to automatically detect passwords, replacing time-consuming manual processes with efficient algorithmic evaluation that maintains high reliability.
3Object-affected harmful factors
If cleartext password detection is implemented, then security can be improved, but false positives may occur
Solution Approach 1:
The system incorporates feedback mechanisms where the probabilistic model continuously refines its assessments based on the context of character sequences. By evaluating the probability of sequences occurring in normal text and comparing against adaptive thresholds, the system receives feedback about potential false positives and adjusts detection sensitivity to maintain high accuracy while reducing false alarms.
Data Source
AI summary
Cleartext passwords represent a security risk. An unencrypted password can be exploited to gain access to a system and/or perform unauthorized functions. This disclosure describes how to detect cleartext passwords in a generalized manner using predictive text classifiers (e.g. Word2Vec). Using a corpus of text, an artificial intelligence model can be built by training a predictive text classifier to identify password anomalies (e.g., areas of text that occur with low statistical probability). Source program code, configuration files, log files, and other types of text can be automatically scanned for cleartext passwords without having to rely on password lists or other limited and/or labor intensive mechanisms, thus improving system security and reducing the chances of data exfiltration and unauthorized actions.