Evolving Ontology Construction from User Content
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to efficiently construct and maintain ontologies from user-generated content on e-commerce websites due to limited domain knowledge, inefficiencies in summarizing concepts, and the evolving nature of data, which leads to challenges in detecting emerging themes and updating semantic structures.
Innovation Solution
A system that utilizes natural language processing, active learning, and semi-supervised learning to automatically detect emerging themes and optimize ontology structures by calculating semantic similarity scores, clustering data entries, and verifying new concepts through a management interface, allowing for minimal human supervision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual ontology construction methods are used, then domain knowledge accuracy is improved, but productivity and efficiency deteriorate due to the large volume and diversity of user-generated content
Solution Approach 1:
The system enables automatic ontology construction through self-service mechanisms where the computing device autonomously receives user-generated content, calculates semantic similarity scores, clusters data entries into themes, detects emerging themes, and updates the ontology database without requiring manual intervention for each step, thereby resolving the contradiction between accuracy and efficiency
Solution Approach 2:
The patent replaces manual mechanical ontology construction with automated computational processes including sentiment analysis, text similarity calculation, syntactic analysis, and theme clustering algorithms that systematically process large volumes of user-generated content to build and update ontologies efficiently while maintaining domain knowledge accuracy
2Device complexity
If traditional summarization methods are used, then concept extraction is simplified, but adaptability deteriorates due to the evolving nature of user-generated content and emerging themes
Solution Approach 1:
The system implements dynamic ontology construction that continuously adapts to evolving user-generated content by periodically receiving new data entries, recalculating semantic similarities, detecting emerging themes through comparison with previous themes, and updating the ontology database to reflect current concepts, thereby achieving both simplified extraction and high adaptability
Solution Approach 2:
The patent incorporates feedback mechanisms where the system compares current themes with previous themes to identify emerging concepts, uses sentiment analysis results to refine concept extraction, and continuously updates the ontology based on analyzed user-generated content, enabling the ontology to evolve dynamically while maintaining manageable complexity
3Measurement precision
If extensive human supervision is applied, then concept verification accuracy is improved, but loss of time increases due to the large volume of data requiring review
Solution Approach 1:
The system applies partial human supervision by automatically performing initial concept extraction and theme detection, then presenting only verified emerging themes and new concepts to human managers for final approval, thereby reducing the time required for human review while maintaining verification accuracy through selective human intervention
Data Source
AI summary
A method and system for constructing an evolving ontology database. The method includes: receiving a plurality of data entries; calculating semantic similarity scores between any two of the data entries; clustering the data entries into a multiple current themes based on the semantic similarity scores; selecting, new concepts from the current themes by comparing the current themes with a plurality of previous themes prepared using previous data entries; and updating the evolving ontology database using the new concepts. The semantic score between any two of the data entries are calculated by: semantic similarity score=Πi=0nsieΣ<sub2>j=0</sub2><sup2>k</sup2>w<sub2>j</sub2>×f <sub2>j</sub2>, where si is weight of features sources, fj is a feature similarity between the two of the data entries, wj is a weight of fj, and j, k and n are positive integers.