Detecting Linguistic Term Evolution via ML Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to detect online shifts in the meaning of linguistic terms, such as pejoration (downgrading) and reappropriation (upgrading), which can lead to harassment and misinterpretation in online environments.
Innovation Solution
The system employs machine learning algorithms and monitoring modules to track the usage and relationships of linguistic terms across multiple messaging sources, identifying shifts in implied meanings over time and attributing changes to specific sources and users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning algorithms and monitoring modules are employed to track linguistic terms across multiple messaging sources, then the ability to detect shifts in implied meanings (pejoration and reappropriation) is improved, but the system complexity and computational resources required increase
Solution Approach 1:
The system segments the monitoring task by dividing it into multiple independent components: individual monitoring modules for different messaging sources, separate machine learning algorithms for different types of linguistic shifts (pejoration vs. reappropriation), and distributed tracking across multiple data sources. This segmentation allows the complex detection task to be broken into manageable parts that can be processed independently, improving detection accuracy without overwhelming system complexity.
Solution Approach 2:
The patent introduces intermediary components including natural language processing layers that mediate between raw messaging data and the detection algorithms, and intermediate representation layers that translate linguistic terms into structured formats for analysis. These intermediaries simplify the interaction between diverse messaging sources and the detection system, enabling accurate detection of meaning shifts while managing computational complexity through standardized processing interfaces.
2Reliability
If the system monitors usage of linguistic terms across multiple messaging sources over time, then the ability to identify emerging forms of harassment and reappropriation attempts is improved, but the time and computational resources required for analysis increase
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing linguistic term usage data as it is collected across messaging sources, organizing it into structured formats with metadata about context, timing, and source. This preliminary organization allows the detection algorithms to query and analyze pre-processed data rather than processing raw messages in real-time, significantly reducing analysis time while maintaining reliable detection of meaning shifts and harassment patterns.
Solution Approach 2:
The monitoring system operates continuously across multiple messaging sources, maintaining ongoing tracking of linguistic term usage rather than performing periodic batch analysis. This continuous monitoring enables the system to detect emerging patterns of pejoration and reappropriation as they develop, improving reliability in identifying harassment attempts while distributing computational load over time to reduce peak analysis requirements.
Data Source
AI summary
The present invention extends to methods, systems, and computer program products for detecting online contextual evolution of linguistic terms. Within messaging sources, some users may actively attempt to (relatively quickly) shift the meaning of a word or term. Some users may attempt to perjorate a word or term to have a more toxic meaning. Other users may attempt to reappropriate a word or term to have a less toxic or even a positive meaning. Aspects of the invention identify shifts in implied meanings of words and/or phrases over time. As such, emerging forms of harassment can be identified more quickly. Aspects of the invention can utilize users' behavioral histories as well as messaging structures to improve confidence when identifying term evolution. Machine learning algorithms can be configured to identify term evolution reducing workload on human moderators.