Cross-Media Event Coreferencing System for Social and News Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in accurately and efficiently extracting event information from diverse media sources like social media and news articles, due to noise, differing formats, and reliability issues, which hinders timely decision-making in fields such as finance, security, and governance.
Innovation Solution
A cross-media event detection and coreferencing system that processes social media and news articles to identify key concepts, filter noise, and verify the credibility of event information, generating coreferenced event representations that combine the strengths of both media types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If social media is used as a source of breaking news, then timeliness of event detection is improved, but reliability of event information deteriorates
Solution Approach 1:
The patent introduces an intermediary verification system that acts as a mediator between social media sources and event detection outputs. The system uses multiple verification mechanisms including cross-source validation, credibility scoring of sources, and automated fact-checking processes to filter and verify social media information before presenting it as confirmed events, thus maintaining timeliness while improving reliability
Solution Approach 2:
The system dynamically adjusts verification parameters and confidence thresholds based on the type of event, source credibility, and available evidence. By changing the stringency of verification parameters adaptively, the system can process social media information quickly for high-confidence events while applying more rigorous verification for ambiguous cases, balancing speed and reliability
2Reliability
If traditional news articles are used for event information, then reliability of event information is improved, but timeliness of event detection deteriorates
Solution Approach 1:
The system performs preliminary monitoring and preprocessing of traditional news sources by maintaining indexes of published articles, extracting key event information in advance, and preparing it for rapid retrieval and cross-validation. This preliminary action allows the system to quickly access verified information from traditional news sources when needed, reducing the effective delay
Solution Approach 2:
The patent merges multiple news sources and social media streams into a unified event detection framework that processes information from both traditional and social media concurrently. By combining sources and using cross-validation, the system can identify events faster through social media while using traditional news articles for verification, achieving both speed and reliability
3Loss of information
If event extraction is performed on multiple media types, then comprehensiveness of event information is improved, but complexity of processing deteriorates
Solution Approach 1:
The system segments the event extraction process into distinct modular components: text ingestion modules for different media types, extraction modules specialized for each format, verification modules, and integration modules. This segmentation allows independent optimization of each component and simplifies the overall system by breaking down the complex task of multi-media processing into manageable, reusable units
Solution Approach 2:
The patent implements universal event schemas and standardized data structures that can represent events from any media type uniformly. The system uses multi-functional processing pipelines that can handle different input formats (social media posts, news articles, reports) through the same core extraction and verification logic, reducing complexity by avoiding separate specialized systems for each media type
4Reliability
If noise filtering is applied to social media postings, then reliability of event information is improved, but loss of useful information deteriorates
Solution Approach 1:
The system applies different filtering strategies to different portions of social media content based on local characteristics. Instead of uniformly filtering all social media posts, the system analyzes each posting's context, source credibility, and content features to apply appropriate filtering intensity. Useful information from credible sources is preserved while noise from unreliable sources is filtered, maintaining both reliability and information completeness
Data Source
AI summary
A method of providing cross-media event linking may include: receiving, at a first input of an event coreferencing system, a stream of social media postings, and at a second input, a stream of news articles; generating, by the event coreferencing system, a first set of event representations representing events referenced by the social media postings, and a second set of event representations representing events referenced by the news articles; determining, by the event coreferencing system, that at least one of the social media postings references a same event referenced by at least one of the news articles, the determining including determining at least one similarity using data of at least one of the first set of event representations corresponding to the at least one of the social media postings and data of at least one of the second set of event representations corresponding to the at least one of the news articles; and transmitting, by an output of the event resolution system to the user system, an alert including at least one coreferenced event representation representing the event referenced by the at least one of the social media postings and the at least one of the news articles.