Automated Text Authenticity Detection via N-gram Shape Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Independent publishers face challenges in distinguishing between legitimate, human-generated content and machine-generated content, which can be of poor quality and misleading, leading to accidental publication and consumer disappointment.
Innovation Solution
A method and system that analyzes textual works by parsing text into N-grams, plotting them in N-dimensional space, transforming to 2-dimensional space, calculating shape descriptors and center of mass, and comparing these features to known human and machine-generated works to determine authenticity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual review of all submitted content is performed, then content quality can be ensured, but time consumption and operational complexity increase significantly
Solution Approach 1:
The system enables content to self-identify its origin through automated analysis of writing patterns, sentence structure, and linguistic features. Machine-generated content automatically reveals its nature through computational detection, eliminating the need for manual verification of each submission.
Solution Approach 2:
The patent replaces the mechanical manual review process with an automated computational system that analyzes textual features, sentence structures, and linguistic patterns to detect machine-generated content, significantly reducing time consumption while maintaining detection accuracy.
2Productivity
If automated detection systems are implemented, then processing speed increases, but system complexity increases
Solution Approach 1:
The detection system is divided into modular components: feature extraction module that identifies linguistic patterns, analysis module that evaluates sentence structures, and detection module that classifies content origin. This segmentation allows each component to be optimized independently while maintaining overall system efficiency.
Solution Approach 2:
The system detects machine-generated content by analyzing changes in multiple textual parameters including sentence length variation, vocabulary diversity, grammatical structure patterns, and semantic coherence metrics. By monitoring deviations in these parameters from human writing norms, the system achieves accurate detection without requiring overly complex algorithms.
3Measurement precision
If comprehensive analysis of textual features is performed, then detection accuracy improves, but computational resources required increase
Solution Approach 1:
The system performs analysis on a selectively chosen subset of textual features rather than exhaustively examining every aspect of the content. By focusing on the most discriminative features such as sentence structure patterns and linguistic stylometry, the system achieves high detection accuracy with reduced computational overhead.
Solution Approach 2:
The system performs preliminary analysis of easily computable features first, such as basic sentence structure and vocabulary patterns. Based on these initial findings, it selectively applies more computationally intensive analysis only to borderline cases, optimizing resource usage while maintaining detection accuracy.
Data Source
AI summary
A system and method for determining whether a textual work submitted for publishing is machine generated or non-machine generated by identifying and quantifying various aspects of the textual work and comparing those aspects to known works. For example, the system and method may identify aspects of a textual work, including, a relationship between the sentences within the textual work, a writing style of the author of the textual work, a grammatical structure of the sentences within the textual work, a quality of the textual work, and other aspects of the textual work. Upon determining that the textual work is machine generated the textual work may be rejected for publishing.