Conversational Interface Generation from Business Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional conversational agent generation techniques require intensive computations and human monitoring, limiting their scope and providing little control for conversation designers, and often behave unpredictably, while also relying on deep learning models that are not suitable for leveraging business domain knowledge.
Innovation Solution
A method that extracts conversational artifacts from business domain knowledge to automatically generate conversational interfaces, using business object models and verbalization files to create intents, entities, and dialog nodes without relying on deep learning or previous conversational logs, allowing for human oversight-free agent creation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If deep learning based language models are used to automate chatbot creation, then the automation extent is improved, but the control for conversation designers over the dialog content and experience deteriorates
Solution Approach 1:
The patent introduces business artifacts (BOM files, VOC files) as intermediary structures that mediate between the automation system and conversation designers. These artifacts serve as structured templates and guidelines that the deep learning model follows, enabling automation while preserving designer control over content and experience through predefined business logic and verification mechanisms
Solution Approach 2:
The patent implements feedback mechanisms where the system provides generated chatbot artifacts to conversation designers for verification and approval. This feedback loop allows designers to review and correct the automated output, ensuring the dialog content and experience meet business requirements while maintaining high automation extent
2Productivity
If deep learning based language models are used to automate chatbot creation, then the productivity is improved, but the reliability deteriorates due to unpredictable behavior
Solution Approach 1:
The patent applies preliminary action by pre-defining business logic, conversation flows, and response templates in BOM and VOC files before the deep learning model generates the chatbot. This preliminary structuring guides the model's behavior and ensures predictable, reliable output that aligns with business requirements, while still benefiting from the productivity advantages of automated generation
3Reliability
If conventional manual design techniques are used for chatbots, then the reliability and control are improved, but the productivity deteriorates due to time consumption
Solution Approach 1:
The patent uses copying by replicating proven business logic patterns and conversation templates from existing BOM and VOC files into the new chatbot generation process. This allows the system to quickly generate reliable chatbots by copying and adapting proven patterns rather than creating everything from scratch, significantly improving productivity while maintaining the reliability and control of manual design through structured templates
Data Source
AI summary
A conversational interface generation method, system, and computer program product that includes determining a conversational artifact for a computer program from a specification of the computer program and generating a conversational interface for the computer program based on the conversational artifact for the computer program included in the specification.