Conversation Controller Dynamic Reply Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional terminal apparatuses for taking orders at restaurants do not allow customers to provide desired information beyond pre-displayed menu options, limiting user interaction and service quality.
Innovation Solution
A conversation controller that outputs reply sentences based on user utterances, using a plan memory unit to store plans with designated next candidate replies, allowing the system to guide users in providing desired information through a series of questions while accommodating discussions on unrelated topics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a terminal apparatus displays fixed menu options for customers to select, then the order taking process is standardized and efficient, but the customer cannot provide desired information beyond the pre-displayed menu options
Solution Approach 1:
The conversation controller dynamically generates and outputs appropriate reply sentences based on the user's utterance content, rather than following a fixed menu selection path. The system adapts its response in real-time to guide the user toward providing desired order information, combining efficiency with flexibility.
Solution Approach 2:
The system implements a feedback mechanism where the conversation controller analyzes the user's utterance and selects appropriate reply sentences from multiple candidates to guide the user. This feedback loop enables the system to steer the conversation toward obtaining necessary order information while accommodating user-initiated topics.
2Loss of information
If the conversation controller outputs multiple candidate reply sentences to guide user input, then the user can be guided to provide desired information, but the system complexity increases
Solution Approach 1:
The system pre-prepares multiple candidate reply sentences for different user utterance types and situations. By having these candidate responses ready in advance, the conversation controller can quickly select and output the most appropriate reply without complex real-time generation, reducing system complexity while maintaining information completeness.
3Stability of the object's composition
If the conversation controller withholds output when the user mentions unrelated topics, then the main conversation flow is preserved, but the user experience deteriorates
Solution Approach 1:
The conversation controller dynamically adjusts its behavior based on the user's input. When the user mentions unrelated topics, the system temporarily withholds the planned reply sentence output and instead responds to the user's actual input, maintaining conversation stability while improving user experience by acknowledging their comments.
Data Source
AI summary
A conversation controller outputs a reply sentence according to a user utterance. The conversation controller comprises a conversation database and a conversation control unit. The conversation database stores a plurality of plans. Each plan has a reply sentence and one or more pieces of next plan designation information for designating a next candidate reply sentence to be output following the reply sentence. The conversation control unit selects one of the plans stored in the conversation database according to a user utterance and outputs a reply sentence which the selected plan has. Then, the conversation control unit selects one piece of the next plan designation information which the plan has according to a next user utterance and outputs a next candidate reply sentence on the basis of the selected piece of the next plan designation information. Some plans have a series of questions as the reply sentences.


