AI News Event Scoring for Insider Trading Actor Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to effectively detect entities that unfairly profit from confidential business information due to the vast amount of complex data generated by the Information Age and the pseudonymity provided by blockchain technology, making it difficult to gather direct evidence of malfeasance.
Innovation Solution
A method and system utilizing AI/ML models to analyze news media data, categorize and score articles based on impact and sentiment, identify anomalies in market valuations, and determine potential bad actors by associating suspicious transactions with non-public information usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data processing techniques are used to analyze news media data, then the system is simple to implement, but it becomes ineffective at detecting bad actors due to the vast amount of complex data
Solution Approach 1:
The patent segments the complex detection task into multiple specialized AI/ML models, each handling specific aspects such as news article analysis, sentiment detection, anomaly identification, and bad actor determination. This segmentation allows each component to be optimized for its specific function while collectively solving the overall detection problem effectively.
Solution Approach 2:
The patent introduces AI/ML models as intermediary components between the raw news media data and the detection outcomes. These intermediary models process and transform the vast amount of complex data into structured insights that can be effectively analyzed, bridging the gap between data volume and detection effectiveness.
2Productivity
If manual analysis of news media data is performed, then the system is simple to operate, but it becomes ineffective due to the immense quantity of data that cannot be processed manually
Solution Approach 1:
The patent replaces manual mechanical analysis with automated AI/ML-based processing systems. These electronic intelligence systems can process vast quantities of news media data at speeds and volumes impossible for human analysts, dramatically increasing productivity while handling the complexity through specialized algorithms.
Solution Approach 2:
The patent changes the operational parameters of data processing by transitioning from human-capable volumes to machine-capable volumes, and from qualitative manual assessment to quantitative algorithmic analysis. This parameter change enables processing of immense data quantities while maintaining systematic control through defined computational parameters.
3Reliability
If cryptocurrency pseudonymity is utilized for transactions, then financial privacy is improved, but it becomes harder to detect bad actors that unfairly use confidential information
Solution Approach 1:
The patent uses AI/ML models as intermediary analysis layers that process transaction data and news media information to infer identities and relationships without requiring direct exposure to sensitive personal information. This intermediary processing maintains detection accuracy while respecting the pseudonymous nature of cryptocurrency transactions.
Solution Approach 2:
The system employs feedback loops where detection results and analyzed patterns are continuously refined through AI/ML learning processes. This feedback mechanism improves detection accuracy over time by learning from confirmed cases while maintaining the ability to operate within the constraints of pseudonymous data.
Data Source
AI summary
A system for detecting bad actors that unfairly profit from confidential information. The system may include a processor and memory that stores instructions that, when executed by the processor, cause the processor to: obtain news articles from at least one news media source; categorize, by pertinent events, each news article according to the pertinent events to which the news articles pertain; score each news article according to a respective impact of each news article; designate, for each of the pertinent events to which the news articles pertain, a corresponding made-public-date; and calculating, for each event of the pertinent events to which the news articles pertain, a corresponding event impact score by aggregating a weighted set of the article impact scores that pertain to the corresponding event.