Deception Detection via Communicative Discourse Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies are inadequate in accurately identifying deceptive or fake textual content, as they fail to effectively analyze the complex rhetorical relationships and communicative actions within text.
Innovation Solution
The use of communicative discourse trees, which represent rhetorical relationships between text fragments, and involve labeling communicative actions to compute a complexity score that indicates the presence of deceptive content by analyzing non-trivial rhetorical relations and nesting levels, with machine learning models for validation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional text analysis methods are used, then the analysis process is simple and fast, but the accuracy of identifying deceptive content is insufficient
Solution Approach 1:
The text is segmented into elementary discourse units (EDUs) that are then organized into a discourse tree structure. This segmentation allows the system to analyze rhetorical relationships at multiple levels (sentence level, paragraph level, document level), improving deception detection accuracy by examining the hierarchical structure of arguments rather than treating text as a flat sequence of words.
Solution Approach 2:
The patent introduces a new dimensional approach by constructing discourse trees that add a hierarchical structure dimension to traditional text analysis. Instead of analyzing only linear text sequences, the system creates a tree-based representation that captures rhetorical relationships, enabling multi-dimensional analysis of argument structure, premise-support relationships, and communicative actions.
2Measurement precision
If comprehensive analysis of rhetorical structures is performed, then deception detection accuracy improves, but computational complexity increases
Solution Approach 1:
The system performs preliminary actions by pre-defining rhetorical relation types (e.g., claim, evidence, counter-argument, premise) and communicative action categories before analysis. This preprocessing of analytical frameworks allows the discourse tree construction to efficiently categorize text elements without requiring complex real-time computations, reducing computational overhead while maintaining comprehensive analysis capability.
Solution Approach 2:
The patent employs parameter changes by adjusting the granularity and depth of discourse tree analysis based on text characteristics. The system can modify analysis parameters such as the level of nesting examined, the types of rhetorical relations prioritized, and the complexity of communicative action labeling, allowing flexible resource allocation that adapts to different text types and deception detection requirements.
Data Source
AI summary
Systems, devices, and methods of the present invention detect deceptive or fake content in text. In an example, a computer system generates, from text a discourse tree that represents rhetorical relationships between fragments of the text. The computer system generates a communicative discourse tree from the discourse tree. The computer system identifies a number of non-trivial rhetorical relations associated with the nonterminal nodes in the communicated discourse tree and, for each terminal edge having a communicative action, a level of nesting of the communicative action. The computer system derives, from the number of non-trivial rhetorical relations and the levels of nesting of the identified communicative actions, a complexity score that is indicative of a level of deception in the text.