Communicative Discourse Trees for Rhetorical Agreement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing keyword-based solutions for autonomous agents fail to accurately determine both topical relevance and rhetorical agreement with user questions, leading to disjointed and inappropriate answers, even when the information is on-topic.
Innovation Solution
The use of communicative discourse trees, which represent rhetorical relationships between sentence fragments, allows for the generation of question and answer trees that can determine the complementarity between questions and candidate answers, ensuring both topical relevance and rhetorical agreement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword-based solutions are used to determine answer relevance, then the system is simple and fast, but the answer quality and rhetorical agreement deteriorate
Solution Approach 1:
The patent segments the answer evaluation process into multiple dimensions: topical relevance assessment and rhetorical agreement assessment. By dividing the evaluation into separate analytical components, the system can independently optimize each aspect without compromising overall performance, thereby achieving both speed and quality.
Solution Approach 2:
The patent introduces discourse trees as an intermediary structure that bridges keyword matching and answer evaluation. These trees represent the rhetorical structure of questions and answers, serving as a mediator that enables sophisticated analysis while maintaining system efficiency through structured representation.
2Measurement precision
If sophisticated discourse analysis is implemented to ensure rhetorical agreement, then answer quality improves, but system complexity increases
Solution Approach 1:
The system segments discourse analysis into hierarchical levels: sentence-level rhetorical relation identification, discourse tree construction, and tree-based comparison. This segmentation allows complex analysis to be performed in manageable stages, reducing overall system complexity while maintaining high answer quality.
Solution Approach 2:
The patent transitions from linear keyword matching to multi-dimensional discourse tree analysis. By representing questions and answers as structured trees with multiple dimensions (rhetorical relations, discourse units, hierarchical levels), the system achieves sophisticated analysis without proportionally increasing complexity.
3Measurement precision
If comprehensive discourse tree analysis is performed for both question and answer, then rhetorical agreement accuracy improves, but processing time increases
Solution Approach 1:
The system performs preliminary action by constructing discourse trees for questions in advance, before answer generation. This pre-processing allows the system to have the question's rhetorical structure ready for rapid comparison with candidate answers, reducing processing time during actual interactions while maintaining high accuracy.
Data Source
AI summary
Systems, devices, and methods of the present invention detect rhetoric agreement between texts. In an example, a rhetoric agreement application accesses a multi-part initial query and generates a question communicative discourse tree that represents rhetorical relationships between fragments of the query. The application identifies a sub-communication discourse tree from the question communicative discourse tree. The application generates a candidate answer communicative discourse tree for each candidate answer of a set of candidate answers. The application computes a level of complementarity between the sub-discourse tree and each candidate answer discourse tree by applying a classification model to the sub-communication discourse tree and candidate answer communicative discourse trees. The application selects an answer from the candidate answers based on the computed complementarity, thereby building a dialogue structure of an interactive session.