Neural Network Credential Pattern Detection and Randomization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional cryptographic key generation systems are vulnerable to pattern recognition by hackers, as they fail to detect and eliminate hidden patterns in credential data, and are susceptible to ransomware and malware attacks, lacking the ability to automatically adapt to password age and eliminate pattern predictability.

Innovation Solution

A method and system utilizing an artificial neural network (ANN) model to detect patterns in transactional credential datasets by correlating historical data, storing them in a graphical embedding storage model, and modifying the patterns to enhance security, making it difficult for unauthorized access.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If rule-based algorithms are used for cryptographic key generation, then encryption can be performed, but the system becomes vulnerable to pattern recognition by hackers

Engineering Contradiction:

Improvedata securityVSAvoidpattern recognition vulnerability

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent replaces traditional rule-based mechanical algorithms with a neural network-based system that learns patterns automatically. The neural network analyzes credential data and generates cryptographic keys without relying on predetermined mathematical rules, making the system adaptive and resistant to pattern recognition attacks.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system dynamically changes parameters based on learned patterns from credential data. Instead of fixed algorithmic rules, the neural network adjusts its internal parameters (weights and biases) to recognize complex patterns in credential data, generating more secure and unpredictable cryptographic keys.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If the same password format or pattern is provided every time, then encryption can be performed efficiently, but hackers can access the password by recognizing the pattern

Engineering Contradiction:

Improveencryption efficiencyVSAvoidsecurity against pattern recognition

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent introduces dynamic pattern recognition and adaptation. The neural network continuously learns from credential data and adapts its pattern recognition capabilities, allowing the system to maintain high encryption efficiency while simultaneously detecting and eliminating predictable patterns that hackers might exploit.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system implements feedback mechanisms where the neural network analyzes encrypted data and credential patterns, learns from the results, and adjusts its behavior accordingly. This feedback loop enables the system to improve its pattern recognition capabilities over time, making it increasingly difficult for hackers to predict passwords.

Inventive Principle:

Principle #23Feedback

3Ease of manufacture

If cryptographic key generation systems use predetermined mathematical rules, then key generation can be performed, but the system gets affected by ransomware and malware

Engineering Contradiction:

Improvekey generation capabilityVSAvoidsusceptibility to malware

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The patent replaces traditional rule-based key generation systems with a neural network-based approach. This substitution makes the system more resilient to malware and ransomware attacks, as the neural network can adapt to new threats and patterns that predetermined rules cannot anticipate.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

4Reliability

If available methods are used for cipher text generation, then encryption can be performed, but the methods lack the ability to detect and eliminate hidden patterns automatically

Engineering Contradiction:

Improveencryption functionVSAvoidautomatic pattern detection capability

Core Design Contradiction:

ReliabilityVSExtent of automation

Solution Approach 1:

The patent implements self-service automation through the neural network, which automatically detects, analyzes, and eliminates hidden patterns in credential data without human intervention. The system autonomously learns from the data and improves its pattern recognition capabilities, maintaining encryption reliability while achieving full automation.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11502817B2Enhanced data security system and method thereof

Publication Date: 2022.11.15 WIPRO LTD

AI summary

The disclosure relates to an enhanced data security system and method thereof. In some embodiments, the method includes receiving the transactional credential dataset from a user application. The transactional credential dataset is provided by a user to the user application. The method further includes storing the transactional credential dataset in nodes of a graphical embedding storage model. The nodes further store historical credential datasets of the user. Further, the method includes determining a correlation among the historical credential datasets using an artificial neural network (ANN) model and detecting a pattern of the transactional credential dataset based on the correlation. The ANN model is trained based on credential datasets provided by users stored in the nodes of the graphical embedding storage model.