Keyword Graph Event Detection in Social Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing event detection methods in social networks face challenges in accuracy due to reliance on word appearance frequency, inability to detect events outside ontologies, and failure to filter malicious intent, leading to unreliable results.
Innovation Solution
A method and system that generate a keyword graph, cluster candidate graphs based on betweenness centrality, and apply user interest as a weight to determine event graphs, enhancing detection accuracy and filtering out malicious content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If event detection is based on word appearance frequency, then the detection process is simple, but the accuracy of event detection is low
Solution Approach 1:
The patent changes the detection parameters from simple word appearance frequency to a multi-dimensional parameter system including user interest scores, social action variations, graph clustering metrics (betweenness centrality), and temporal patterns. This transforms the detection approach from univariate to multivariate analysis, significantly improving accuracy while maintaining systematic processing.
Solution Approach 2:
The patent creates a composite detection framework that combines multiple data sources and analysis methods: keyword graphs, user interest profiles, social action data, ontology relationships, and clustering algorithms. This composite approach integrates diverse elements to achieve higher detection accuracy than any single method could provide alone.
2Measurement precision
If ontology-based graph detection is used, then event clarity is improved, but only events included in the ontology can be detected
Solution Approach 1:
The patent segments the event detection process into multiple independent modules: keyword graph construction, user interest calculation, social action analysis, ontology-based verification, and clustering. This segmentation allows the system to handle both ontology-known events and novel events separately, maintaining clarity for structured events while capturing versatility for unstructured events.
Solution Approach 2:
The patent introduces keyword graphs as an intermediary structure that bridges ontology-based detection and free-text analysis. The graphs serve as a mediator that can incorporate both ontology-defined relationships and emergent patterns from social data, allowing the system to maintain structural clarity while adapting to diverse event types.
3Measurement precision
If user interest is applied as a weight in graph clustering, then event detection accuracy is enhanced, but the complexity of the detection system increases
Solution Approach 1:
The patent performs preliminary calculations of user interest scores and social action variations before the main event detection process. By pre-computing these metrics and storing them as weights in the keyword graphs, the system reduces the complexity of the main detection algorithm while maintaining enhanced accuracy through the pre-prepared user interest information.
4Productivity
If traditional keyword listing is used for event detection, then the result is simple to generate, but users cannot readily refer keywords to events
Solution Approach 1:
The patent transforms the flat keyword list into a multi-dimensional keyword graph structure where keywords are nodes connected by semantic and contextual relationships. This dimensional transformation preserves the simplicity of keyword-based output while adding structural context that enables users to trace keywords back to their source events through the graph's hierarchical and relational properties.
Data Source
AI summary
Provided is a method and system for detecting a graph-based event in a social network, the method including generating a keyword graph based on social data collected from a social network system; separating the keyword graph into a plurality of candidate graphs through clustering; calculating a user interest for each of the plurality of candidate graphs based on a per-time variation of a social action associated with the social data; and determining k candidate graphs from among the plurality of candidate graphs as an event graph based on the user interest. Here, k denotes a natural number.