ML Data-Origin Link Determination for Named Entities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining data-origin links between two named entities based on user input data and inheritance information is challenging due to inaccuracies in user-submitted data and computational challenges associated with large-scale databases.
Innovation Solution
A system that uses a computing device to receive datasets for two named entities, identify data matches, extract features, and input these features into a machine learning model to determine data-origin links.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If user-submitted data is used to determine data-origin links, then the process is simple and fast, but the accuracy is low due to inaccuracies in user input
Solution Approach 1:
The patent introduces an intermediary verification mechanism that compares user-submitted data against independently verified data from multiple sources (social media platforms, news outlets, academic databases). This intermediary layer acts as a mediator between the simple user input and the accurate final determination, resolving the contradiction by maintaining processing speed while improving accuracy through cross-validation.
Solution Approach 2:
The patent replaces the mechanical verification process (manual checking of user data) with an automated machine learning system that uses natural language processing and pattern recognition algorithms. This substitution maintains simplicity and speed while dramatically improving accuracy by automatically detecting inconsistencies and verifying data against multiple sources.
2Measurement precision
If large-scale databases are searched to verify data-origin links, then the accuracy is improved, but the computational complexity and time consumption increase significantly
Solution Approach 1:
The patent segments the large-scale database search process into hierarchical levels: first searching within the user's own data tree, then expanding to connected data trees, and finally searching broader databases only when necessary. This segmentation reduces computational complexity by avoiding unnecessary searches of entire large databases while maintaining accuracy through systematic expansion of the search scope.
Solution Approach 2:
The patent performs preliminary filtering and preprocessing of data before conducting full database searches. By first identifying potential matches using quick comparison algorithms and only then conducting detailed verification searches, the system reduces computational complexity while maintaining accuracy through staged verification processes.
3Measurement precision
If comprehensive data comparison is performed across large datasets, then the accuracy of identifying data-origin links is improved, but the memory usage and processing time increase
Solution Approach 1:
The patent extracts and compares only the essential identifying features and data segments relevant to origin determination, rather than performing comprehensive comparison of entire datasets. By extracting key identifiers, timestamps, and unique data characteristics for comparison while ignoring redundant information, the system improves accuracy of matching while significantly reducing processing time and memory usage.
4Reliability
If repeated comparison of massive amounts of database data is performed, then thorough verification is achieved, but network bandwidth is consumed and computational efficiency decreases
Solution Approach 1:
The patent applies local quality verification by focusing computational resources on specific segments of data that require verification rather than uniformly processing all data. By identifying and prioritizing verification of critical data segments (those with higher uncertainty or importance) while using simpler verification for well-established data, the system achieves thorough verification where needed while maintaining computational efficiency overall.
Data Source
AI summary
A system includes a computing device having one or more processors and memory configured to store instructions, wherein the instructions, when executed by the one or more processors, cause the one or more processors to perform steps including receiving a first named entity dataset associated with a first named entity, receiving a second named entity dataset associated with a second named entity who is a potential named entity linked to the first named entity, identifying a plurality of data matches, each data match being a named entity whose data segments match the first named entity dataset, the second named entity dataset, or both, extracting features from the plurality of data matches, and inputting the extracted features into a machine learning model to determine a data-origin link between the first named entity and the second named entity.