Device Classification Service Rule Reuse via Clustering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In computer networks, especially in IoT environments, the dynamic nature of device connections leads to cumbersome and conflicting device classification rules, resulting in misclassifications and poor accuracy when either new rules are generated frequently or existing rules are always reused.

Innovation Solution

A device classification service that forms clusters of endpoint devices based on their attributes and applies initial classification rules, computes metrics to determine whether to associate existing rules or generate new ones, leveraging a central rule database and machine learning to decide when to reuse or create new rules.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If new classification rules are generated frequently to adapt to network changes, then adaptability improves, but rule complexity and conflicts increase

Engineering Contradiction:

ImproveadaptabilityVSAvoidrule complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system merges classification rules across multiple networks by identifying equivalent rules in a central repository and reusing them, rather than generating new rules independently in each network. This consolidation reduces overall rule complexity while maintaining adaptability through centralized rule management and sharing.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The central rule database serves multiple networks simultaneously, providing a universal repository of classification rules that can be reused across different network contexts. This multi-functional approach allows a single rule to serve multiple purposes and networks, reducing redundant rule generation and complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Device complexity

If existing classification rules are always reused to reduce complexity, then rule complexity decreases, but classification accuracy deteriorates

Engineering Contradiction:

Improverule complexityVSAvoidclassification accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The system dynamically decides whether to reuse or generate new rules based on network-specific conditions and characteristics. Rather than statically reusing all existing rules, the system adapts its rule selection and generation strategy based on real-time network behavior analysis, ensuring both low complexity and high accuracy.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system uses feedback from network behavior observation to evaluate the performance of reused rules and determine when new rules should be generated. By monitoring classification accuracy and network patterns, the system adjusts its rule reuse strategy, generating new rules only when necessary to maintain accuracy while minimizing unnecessary complexity.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If classification rules are generated for each network independently, then classification accuracy improves, but maintenance difficulty increases

Engineering Contradiction:

Improveclassification accuracyVSAvoidmaintenance difficulty

Core Design Contradiction:

Measurement precisionVSEase of repair

Solution Approach 1:

The system adds a new dimension to rule management by introducing a centralized repository layer above individual network rule sets. This additional organizational dimension allows rules to be maintained centrally while still supporting network-specific accuracy requirements, effectively separating maintenance concerns from deployment concerns.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

4Ease of repair

If a central rule database is implemented to share rules across networks, then maintenance ease improves, but system complexity increases

Engineering Contradiction:

Improvemaintenance easeVSAvoidsystem complexity

Core Design Contradiction:

Ease of repairVSDevice complexity

Solution Approach 1:

The system introduces a central rule database as an intermediary layer between individual networks and their classification rules. This intermediary provides centralized maintenance capabilities while using standardized interfaces and protocols to minimize the complexity burden on individual network systems, allowing maintenance improvements without proportional complexity increases.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS10999146B1Learning when to reuse existing rules in active labeling for device classification

Publication Date: 2021.05.04 CISCO TECHNOLOGY INC

AI summary

In various embodiments, a device classification service forms a device cluster by applying clustering to attributes of endpoint devices observed in one or more networks. The device classification service applies an initial device classification rule to the endpoint devices in the device cluster, based on one or more of the endpoint devices in the device cluster matching the initial device classification rule. The device classification service computes metrics for the initial device classification rule that quantify how well the attributes of the endpoint devices in the device cluster match the initial device classification rule. The device classification service decides, based on the metrics, whether to associate the initial device classification rule with the device cluster or generate a new device classification rule based on the device cluster.