Directed Graph to Taxonomy Conversion with Analyst Guidance

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current algorithms for converting directed graphs to spanning trees, such as rooted and directed spanning trees, cannot effectively utilize human judgment or preferences, making it difficult to manage large graphs efficiently.

Innovation Solution

The method involves pruning paths in a directed graph using a branch optimization algorithm like the Edmunds algorithm, incorporating human-specified subtree preferences and recommendations to guide the conversion, allowing for human insight in the pruning process and weighting edges based on these preferences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If automated algorithms (Edmonds/Chu-Liu/Edmunds) are used to convert directed graphs to spanning trees, then the conversion speed and automation level are improved, but the ability to incorporate human judgment and preferences is lost

Engineering Contradiction:

Improveautomation of graph to spanning tree conversionVSAvoidability to incorporate human judgment

Core Design Contradiction:

Extent of automationVSAdaptability or versatility

Solution Approach 1:

The patent introduces an intermediary system that bridges automated algorithms and human judgment. The system presents multiple candidate spanning trees generated by algorithms to analysts, who then provide feedback and preferences. This intermediary process allows both automated speed and human insight to coexist in the conversion workflow.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback loops where analyst preferences and manual adjustments to candidate spanning trees are fed back into the algorithm. This feedback mechanism allows the automated system to learn from and adapt to human judgment, continuously improving the quality of conversions while maintaining automation.

Inventive Principle:

Principle #23Feedback

2Adaptability or versatility

If manual transformation of directed graphs to spanning trees is performed, then human judgment and preferences can be incorporated, but the process becomes too tedious and impossible for large graphs

Engineering Contradiction:

Improveability to incorporate human judgmentVSAvoidconversion speed

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

Instead of requiring complete manual transformation of the entire graph, the system applies partial automation by generating candidate spanning trees algorithmically and then applying human judgment only to select and refine the best options. This partial action approach maintains productivity while incorporating necessary human insight.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The automated algorithms perform preliminary action by generating multiple candidate spanning trees before human analysis. This preliminary computational work reduces the scope of manual effort required, as analysts only need to evaluate and select from pre-generated candidates rather than creating the spanning tree from scratch.

Inventive Principle:

Principle #10Preliminary action

3Manufacturing precision

If existing algorithms prioritize structural optimization, then mathematical optimality is achieved, but meaningful relationships and preferences requiring human insight are ignored

Engineering Contradiction:

Improvemathematical optimality of spanning treeVSAvoidloss of meaningful relationships

Core Design Contradiction:

Manufacturing precisionVSLoss of information

Solution Approach 1:

The system changes the parameters of optimization by incorporating multiple objective functions beyond pure structural metrics. Analyst preferences, domain-specific constraints, and relationship weights are transformed into adjustable parameters that the algorithm optimizes alongside mathematical criteria, preserving meaningful relationships while achieving optimality.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent creates a composite optimization approach that combines multiple types of information: mathematical structure, domain knowledge, analyst preferences, and relationship weights. This composite methodology integrates diverse information sources into a unified spanning tree selection process that achieves both mathematical optimality and preservation of meaningful relationships.

Inventive Principle:

Principle #40Composite materials

Data Source

PatentUS9092547B2Transforming a graph to a tree in accordance with analyst guidance

Publication Date: 2015.07.28 WALMART APOLLO LLC

AI summary

Methods are disclosed for converting a directed graph to a taxonomy using guidelines from a user. An initial tree is output from a first pruning step in which subtree preferences (and other weights) are applied to preserve or remove paths from a node to one or more levels of descendent nodes. Subtree preferences (and infoboxes) may specify rules for automatically generating recommendations during application to nodes. In a second pruning step, the directed graph is again processed with additional weightings applied to edges in the graph in accordance with the recommendations. The recommendations may be human defined. Recommendations may specify a recommended ancestor for a particular node and may include a weighting to be applied to the recommendation itself, if there are multiple conflicting recommendations for the same node. Recommendations may also specify what standard weight to apply to the edge of the best parent.