Subgraph Optimization with Multi-Objective Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for solving non-trivial multi-objective optimization problems, such as choosing an optimal subgraph, face challenges due to conflicting objective functions and an infinite number of possible solutions, particularly in linear optimization which loses the makeup of elements chosen and assumes benefits are independent of other elements.
Innovation Solution
A system that uses machine learning to classify and bundle data into subgraphs with features, creating subfeature and metafeature scores to evaluate and compare subgraphs, allowing for the selection of optimized subgraphs based on multiple criteria without relying on integer programming or single-score optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If single objective function optimization (linear optimization) is used to find the best subgraph selection, then the optimization process is simple and fast, but it suffers severe defects by losing the makeup of elements chosen and assuming benefits are independent of other elements
Solution Approach 1:
The patent segments the single objective function into multiple objective functions, each representing a different aspect of subgraph quality (e.g., coverage, diversity, importance). Instead of optimizing one aggregate score, the system evaluates multiple separate objectives to preserve information about different compositional aspects of the subgraph.
Solution Approach 2:
The patent transitions from a single-dimensional optimization (one objective function) to a multi-dimensional evaluation space where each dimension represents a different objective. This allows the system to maintain information about element composition across multiple dimensions rather than collapsing it into a single score.
2Ease of operation
If manual selection of subgraphs is performed on an ad hoc basis, then the user can exercise judgment and context, but the process is tedious and time-consuming for non-trivial multi-objective optimization problems
Solution Approach 1:
The patent performs preliminary computation by pre-evaluating multiple objective functions for all candidate subgraphs before the user needs to make a selection. This preparatory work creates a structured set of evaluated options that the user can then efficiently compare, reducing the time required for manual selection while preserving user judgment.
Solution Approach 2:
The patent introduces an intermediary computational layer that automatically evaluates and scores subgraphs across multiple objectives. This intermediary system processes the complex multi-objective analysis, presenting refined options to the user that maintain the benefits of user judgment while eliminating the tedium of manual evaluation.
3Reliability
If conventional methods are used for non-trivial multi-objective optimization problems, then the solution approach is traditional and well-understood, but the objective functions are conflicting and the number of possible solutions approaches infinity
Solution Approach 1:
The patent applies partial action by selecting and evaluating only the most relevant objective functions for the specific problem context, rather than attempting to evaluate all possible objectives. This selective approach reduces the effective solution space from infinity to a manageable subset of meaningful trade-offs.
Solution Approach 2:
The patent changes the parameters of the optimization problem by transforming the infinite solution space into a discrete set of candidate subgraphs with explicitly evaluated objective values. This parameter transformation allows conventional optimization techniques to be applied effectively to a reduced, well-defined problem space.
Data Source
AI summary
A subgraph storage memory stores subgraphs populated with data classified into features based on data title; each of the subgraphs has a plurality of elements; each of the elements has a features with different values. For each subfeature of a plurality of subfeatures, the different values assigned to a combination of one of the elements and one or more features are used to create a method to evaluate the subgraphs on the one or more features, as a subfeature which provides scoring information on a subgraph. For each metafeature, plural subfeatures among the plurality of subfeatures are selected to be composited into the metafeature; and for each subgraph, the plural subfeatures of each metafeature are composited to provide metafeature scores. Selected subgraphs in a universe of subgraphs with metafeature scores are presented as context for comparing the selected subgraphs; each of the selected subgraphs has different selections of elements.