Virtual Persuasive Dialogue Using Communicative Discourse Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous agents are limited in presenting information effectively to users, as they often rely on monologues and lack the ability to engage in persuasive dialogues that mimic human conversations, failing to provide richer discourses and effectively adjust user opinions.
Innovation Solution
The development of a computer-implemented method for creating virtual persuasive dialogues by analyzing electronic textual sources, identifying argumentation, and presenting content in a dialogue format, using communicative discourse trees and machine-learning techniques to generate and structure utterances between virtual actors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If autonomous agents use traditional monologue format to present information, then information delivery is simple and direct, but the ability to engage users and adjust user opinions is limited
Solution Approach 1:
The patent segments information delivery into multiple virtual actors representing different viewpoints, each delivering arguments in a structured dialogue format. This segmentation enables the system to present diverse perspectives and engage users more effectively while managing complexity through modular discourse tree structures.
Solution Approach 2:
The patent introduces virtual actors as intermediaries between the information source and the user. These virtual actors embody different viewpoints and engage in persuasive dialogue, serving as mediators that make information delivery more engaging and adaptable to user opinions without requiring the system itself to be directly complex.
2Loss of information
If autonomous agents present information in detailed dialogue format, then user engagement and information richness improve, but information delivery time and processing complexity increase
Solution Approach 1:
The patent performs preliminary action by pre-structuring information into discourse trees that identify rhetorical relationships and argumentation patterns before dialogue generation. This preprocessing enables faster retrieval and assembly of dialogue content, maintaining information richness while reducing generation time.
Solution Approach 2:
The patent changes parameters by transforming static text into structured discourse representations with labeled rhetorical relationships. This parameter transformation enables efficient manipulation and retrieval of information elements during dialogue generation, balancing detail delivery with processing speed.
3Productivity
If autonomous agents use simple information presentation, then processing speed is fast, but the ability to provide richer discourses and engage users is reduced
Solution Approach 1:
The patent extracts key rhetorical relationships and argumentation structures from source text into discourse trees, separating essential discourse elements from full text content. This extraction enables efficient processing of core information while preserving discourse richness through structured representation of rhetorical relationships.
Data Source
AI summary
Techniques are disclosed for generating a virtual persuasive dialogue. In an example, a dialogue application receives a selection of a topic from a user device. The application identifies document results that are associated with the topic. Using communicative discourse trees, the application identifies document results that include argumentation, transforms these document results into a dialogue form, and presents the results to a user device as a virtual persuasive dialogue.