Keyword Expertness via Category Dispersion in Knowledge Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for re-ranking search results do not effectively determine the expertness level of a target keyword based on the dispersion of categories associated with relevant documents, which is crucial for user comprehension and material selection.
Innovation Solution
A computer-implemented method that prepares a document set associated with the target keyword, identifies categories for each document, and determines the expertness level by calculating the degree of dispersion among these categories using a Directed Acyclic Graph (DAG) to assess the keyword's usage across different categories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current re-ranking techniques are used, then search results can be re-ordered, but the expertness level of target keywords cannot be accurately determined based on category dispersion
Solution Approach 1:
The patent introduces a category dispersion measurement mechanism as an intermediary to bridge the gap between document categories and keyword expertness. By measuring how dispersed categories are around a keyword in the knowledge graph, the system accurately determines expertness levels without losing category information.
Solution Approach 2:
The patent replaces traditional mechanical re-ranking mechanisms with a knowledge graph-based semantic analysis system. This substitution enables accurate expertness determination by utilizing the structural relationships and category dispersion in the knowledge graph rather than conventional search algorithms.
2Adaptability or versatility
If category dispersion is considered for expertness determination, then user comprehension can be improved, but the system complexity increases
Solution Approach 1:
The knowledge graph serves as a universal structure that performs multiple functions: it stores category information, measures dispersion, and determines expertness levels. This multi-functionality reduces system complexity by consolidating what would otherwise require separate systems into a single integrated knowledge graph.
Solution Approach 2:
The system uses the knowledge graph's inherent structural properties to automatically determine expertness levels without requiring external complex processing. The category dispersion measurement leverages the graph's existing topology, allowing the system to self-determine expertness based on natural relationships within the graph.
Data Source
AI summary
A method, a computer system, and a computer program product for determining an expertness level for a target keyword. The method includes preparing a document set associated with the target keyword, identifying, for each document in the document set, a category to which the document in the document set belongs, and determining an expertness level for the target keyword, based on a degree of dispersion of the identified categories. Each document in the document set may be associated with a category, and a graph which defines relations among the identified categories is generated from the identified categories to obtain the degree of dispersion of the identified categories. On condition that the degree of dispersion is lower, the expertness level for the target keyword may be determined to be higher or a higher expertness level is assigned to the target keyword.