AI Entity Classification Taxonomy Framework
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing entity classification methods using artificial intelligence models face challenges such as instability in classification results, lack of professional knowledge, and inability to manage complex real-world customer requirements, leading to inefficient network security measures.
Innovation Solution
The implementation of an AI classification model using a defined taxonomy framework, where the classification system iteratively generates prompts to the AI model based on device properties, allowing for accurate and granular classification of entities within a network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing AI models are used for entity classification, then classification can be performed, but the classification results are unstable and lack reliability
Solution Approach 1:
The patent applies preliminary action by pre-defining a comprehensive taxonomy framework with hierarchical categories before classification occurs. This pre-structured framework guides the AI model through predetermined classification paths, ensuring stable and reliable results while reducing the need for time-consuming taxonomy management adjustments during operation.
Solution Approach 2:
The system implements feedback mechanisms where classification results are validated against the predefined taxonomy framework. If inconsistencies arise, the system can adjust or request reclassification, ensuring reliability. The feedback loop also allows continuous refinement of classification decisions without requiring manual taxonomy restructuring.
2Adaptability or versatility
If existing AI models are used for entity classification, then classification can be performed, but the models lack professional knowledge and cannot handle complex real-world requirements
Solution Approach 1:
The patent segments the classification task into hierarchical levels (e.g., device type, function, protocol, vendor) within the taxonomy framework. This segmentation allows the AI model to handle complex device properties systematically by classifying one attribute at a time, improving adaptability while managing complexity through structured decomposition.
Solution Approach 2:
The predefined taxonomy framework serves multiple functions: it provides classification structure, validates AI outputs, guides prompt generation, and ensures consistency across different classification scenarios. This multi-functionality enhances the model's ability to handle complex requirements without proportionally increasing management complexity.
3Productivity
If manual taxonomy management is performed, then classification accuracy can be maintained, but time and labor requirements increase significantly
Solution Approach 1:
The system enables self-service by using the AI model to automatically perform classification tasks based on the predefined taxonomy framework. The model generates its own prompts and makes classification decisions autonomously, eliminating the need for manual taxonomy management while maintaining accuracy. The framework itself serves the system's classification needs without human intervention.
Solution Approach 2:
The patent replaces manual mechanical taxonomy management with an automated AI-based system. The AI model processes device information, generates classification prompts, and assigns categories automatically, substituting human labor with intelligent automation. This maintains classification accuracy while dramatically improving productivity and reducing operational effort.
Data Source
AI summary
Systems and methods for entity classification via an artificial intelligence model using a defined taxonomy framework are described. Entity classification includes generating a first query for a classification model, the first query including a first set of options for classification of an entity at a first classification granularity level of a taxonomy framework, providing the first query comprising the first set of options for classification to the classification model, receiving, from the classification model, a selection of one or more options of the first set of options for classification, and determining a classification of the entity based at least in part on the selection of the one or more options of the first set of options.