Device Attribute Profiling with Machine Learning for Conflicting String Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cybersecurity solutions struggle to accurately identify device attributes due to the lack of a uniform standard for querying and obtaining device information, leading to inconsistent and conflicting data from various sources, which complicates profiling and securing network environments.
Innovation Solution
Utilizing machine learning models, specifically neural networks with an inner product layer, to analyze string field conventions in device data for consistent and accurate device attribute identification, employing a supervised learning process with labeled training data sets to improve device profiling and security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If device data is obtained from various sources without a uniform standard, then more device information can be collected, but the data becomes inconsistent and conflicting making accurate device attribute identification impossible
Solution Approach 1:
The patent transforms device identification from exact matching of device attributes to probabilistic matching of string field conventions. Machine learning models output confidence scores indicating the likelihood that a device matches a particular type, allowing the system to work with incomplete or conflicting data from multiple sources without requiring perfect accuracy from any single data point
Solution Approach 2:
The patent introduces machine learning models as an intermediary layer between raw device data and device attribute identification. These models learn string field conventions from training data and use them to interpret and reconcile conflicting information from multiple data sources, transforming unstructured device strings into reliable device attribute predictions
2Device complexity
If traditional device profiling methods are used without machine learning, then the system is simpler to implement, but it cannot accurately identify device attributes from non-standardized data
Solution Approach 1:
The patent performs preliminary training of machine learning models on labeled device data before deployment. This pre-training phase allows the models to learn string field conventions and patterns from diverse data sources, so that when the system is deployed, it can immediately begin accurately identifying device attributes without requiring complex real-time analysis or manual intervention
Solution Approach 2:
The patent replaces traditional mechanical or rule-based device profiling systems with machine learning-based probabilistic models. Instead of using fixed rules or exact matching algorithms, the system uses trained neural networks that can generalize from training data and handle the variability and inconsistency inherent in non-standardized device data from multiple sources
Data Source
AI summary
A system and method for identifying device attributes based on string field conventions. A method includes applying at least one machine learning model to an application data set extracted based on a string indicated in a field of device data corresponding to a device, wherein each of the at least one machine learning model is trained based on a training data set including a plurality of second strings and a plurality of device attribute labels, wherein each device attribute label corresponds to a respective second string of the plurality of second strings, wherein each of the at least one machine learning model is configured to output a predicted device attribute for the device based on the first string; and identifying, based on the output of the at least one machine learning model, a device attribute of the device.