Feature Relationship Recommendation via Knowledge Graph Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for recommending feature relationships in ALIPAY's security environment are inefficient due to reliance on manual processes and brute-force searches, which consume significant resources and fail to effectively utilize service knowledge to find valid relationships.
Innovation Solution
A feature relationship recommendation method that involves determining parameter instances from service data, constructing a knowledge graph based on these instances and operator entities, and extracting new relationships using an inference rule, such as a path ranking algorithm, to recommend newly added feature relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual search, selection, and determining methods are used to recommend feature relationships, then service knowledge can be utilized, but the process depends heavily on manpower and is inefficient
Solution Approach 1:
The patent introduces a heuristic search algorithm as an intermediary between manual operations and automated processing. The algorithm acts as a mediator that guides the search process through evaluation models, reducing the burden on manual operations while maintaining the ability to utilize service knowledge. This resolves the contradiction by partially automating the process through the heuristic intermediary.
Solution Approach 2:
The patent replaces pure manual mechanical operations with a hybrid system combining heuristic algorithms and evaluation models. The automated heuristic search substitutes for manual searching, while the evaluation model assists manual selection, thereby improving productivity without completely eliminating the ability to utilize service knowledge.
2Reliability
If brute-force search is used to find feature relationships, then comprehensive search can be performed, but it consumes too many resources
Solution Approach 1:
The patent changes the search parameters by introducing heuristic functions and evaluation models that guide the search process. Instead of uniformly exploring all possibilities (brute-force), the search parameters are dynamically adjusted based on heuristic evaluations, reducing computational resource consumption while maintaining reliable search results through informed parameter selection.
Solution Approach 2:
The patent applies partial action by using heuristic search to explore only the most promising portions of the search space rather than performing exhaustive brute-force search. The evaluation model identifies and focuses computational resources on relevant feature relationships, achieving sufficient reliability without excessive resource consumption.
3Extent of automation
If heuristic search algorithm is used to perform search, then automated search can be performed, but service knowledge is not introduced and search space remains large
Solution Approach 1:
The patent segments the search process into two distinct phases: automated heuristic search for exploring the search space, and manual or assisted evaluation for selecting feature relationships. This segmentation allows automated processing to handle the large search space while service knowledge is applied in the evaluation phase, preventing information loss.
Solution Approach 2:
The patent performs preliminary automated heuristic search to identify candidate feature relationships before applying service knowledge through evaluation models. This preliminary action reduces the search space while maintaining automation, and subsequent evaluation ensures service knowledge is properly utilized in selecting the final relationships.
4Extent of automation
If heuristic search with evaluation model is used, then automated selection can be performed, but it is difficult to find valid feature relationships
Solution Approach 1:
The patent implements feedback mechanisms where evaluation models assess candidate feature relationships and provide feedback to the heuristic search algorithm. This feedback loop allows the system to learn from previous evaluations and adjust the search strategy, improving the ability to find valid feature relationships while maintaining automated selection.
Solution Approach 2:
The patent enables the system to self-improve by using evaluation results to refine the heuristic search parameters and evaluation criteria. The automated system serves itself by continuously optimizing its search and selection processes, thereby improving feature relationship validity without requiring constant manual intervention.
Data Source
AI summary
Disclosed herein are methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating relationship recommendations. One of the methods includes: determining, by a computing device, parameter entities that comprise event entities associated with node entities; constructing, by the computing device, a knowledge graph based on relationships between the parameter entities and predetermined operator entities; extracting, by the computing device, a newly added relationship from the knowledge graph based on an inference rule; and providing, by the computing device, a relationship recommendation corresponding to the newly added relationship.