Node Selection in Data Trees Using Interaction History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying related nodes in large-scale databases, such as genealogical databases, is challenging due to the sheer amount of data and the computational infeasibility of comparing numerous datasets without a concrete strategy.
Innovation Solution
A computing system that continuously updates a set of nodes addable to a data tree by using a machine learning model to select candidate nodes based on past user interactions, tracking recently interacted nodes, and presenting them in a graphical user interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If users manually identify and compare datasets to find related nodes, then they can build accurate data trees, but the process becomes computationally infeasible and time-consuming due to the large volume of data
Solution Approach 1:
The patent introduces an intermediary system consisting of machine learning models and domain knowledge graphs that mediate between the user and the large-scale database. The system automatically identifies candidate nodes and their relationships by processing user interactions and querying the database, thereby eliminating the need for users to manually compare datasets while maintaining high accuracy in node identification
Solution Approach 2:
The patent replaces the mechanical manual process of dataset comparison with automated computational systems. Machine learning models analyze user interaction patterns, and domain knowledge graphs encode relationships between data entities, automatically identifying related nodes without requiring users to perform time-consuming manual comparisons
2Adaptability or versatility
If the system presents all candidate nodes to users, then users have complete information for decision-making, but the interface becomes overwhelming and difficult to navigate
Solution Approach 1:
The patent segments the large set of candidate nodes into manageable subsets based on user interaction history and domain knowledge. The system divides nodes into different categories or priority levels, presenting only the most relevant candidates to users at any given time, thereby maintaining completeness of options while improving interface usability
Solution Approach 2:
The patent applies local quality by customizing the presentation of candidate nodes based on specific user contexts and interaction patterns. Different users or different stages of interaction receive differently curated sets of nodes, with the system adapting the level of detail and type of information presented to match the user's current needs and expertise
3Ease of manufacture
If the system uses simple random selection for candidate nodes, then the implementation is straightforward, but the relevance and quality of recommended nodes deteriorates
Solution Approach 1:
The patent introduces machine learning models and domain knowledge graphs as intermediaries between simple random selection and the final candidate node recommendations. These intermediaries process user interaction data and domain knowledge to automatically identify and rank relevant nodes, thereby maintaining implementation feasibility while dramatically improving recommendation quality
Solution Approach 2:
The patent changes the parameters of node selection from uniform random probability to probability distributions based on user interaction history and domain knowledge. The system dynamically adjusts selection parameters such as relevance weights, interaction frequency, and domain-specific relationship strengths to optimize the quality of recommended nodes while maintaining computational efficiency
Data Source
AI summary
A computing server may continuously update a set of nodes that are addable to a data tree based on past interactions of the user with one or more nodes. The computing server may track a recently interacted set of interacted nodes with which the user has interacted within a number of past interactions. The computing server may select a pool of candidate nodes based on the recently interacted set. At least one of the candidate nodes is within a domain boundary of one of the interacted nodes that is in the recently interacted set. The domain boundary may be determined by the degree of relationship. The computing server may present one or more candidate nodes in the pool as a version of the continuously updated set of nodes. The computing server may update the pool of candidate nodes as additional interactions performed by the user updates the recently interacted set.