Configurable Conversation Engine for Dynamic Chatbot Customization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional chatbots are cumbersome and error-prone due to their hardcoded scripts, making modifications difficult and inefficient for customizing interactions with users.
Innovation Solution
A customizable conversation engine that receives user utterances, determines attributes, and performs actions based on a conversation engine configuration and chatbot configuration, allowing for dynamic task execution and entity management across multiple chatbots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional chatbots use hardcoded scripts to specify actions, then the chatbot can perform specific tasks, but the chatbot becomes cumbersome and error-prone when modifications are needed
Solution Approach 1:
The patent implements dynamic configuration files that allow chatbot behaviors to be modified without changing the underlying code. The conversation engine reads configurations from external files (XML, JSON, YAML, etc.), enabling runtime adjustments to chatbot personality, responses, and interactions. This dynamic approach resolves the contradiction by making the system adaptable while keeping modification simple through file editing rather than code changes.
Solution Approach 2:
The patent segments the chatbot system into distinct components: the conversation engine (core processing logic), configuration files (behavior definitions), and external data sources. This segmentation allows modifications to be made in isolated configuration files without affecting the core engine, reducing complexity and errors while maintaining full customizability of chatbot behaviors.
2Adaptability or versatility
If chatbots are customized for specific tasks, then they can perform specialized functions, but the process of customization becomes error-prone and difficult
Solution Approach 1:
The patent introduces configuration files as an intermediary layer between the conversation engine and task-specific requirements. These files serve as a safe, structured interface for customization, allowing task-specific behaviors to be defined without directly modifying engine code. This intermediary approach reduces errors by providing validation, templates, and a clear separation between core logic and customizable parameters.
Solution Approach 2:
The conversation engine automatically loads and applies configurations from external files without requiring manual programming for each task. The system self-adjusts its behavior based on the provided configuration, reducing human error in customization while maintaining task-specific functionality. The engine handles parsing, validation, and application of configurations automatically.
3Ease of operation
If chatbots use natural language processing, then user interaction improves, but the system requires configurable states and actions that increase complexity
Solution Approach 1:
The patent implements a universal configuration structure that handles multiple conversation states, user inputs, and system actions through a single standardized framework. The configuration files use consistent syntax and structure across different chatbots and tasks, allowing the same engine to handle diverse natural language interactions without requiring separate complex configurations for each scenario. This universal approach maintains ease of operation while managing complexity through standardization.
Data Source
AI summary
A conversation engine performs conversations with users using chatbots customized for performing a set of tasks that can be performed using an online system. The conversation engine loads a chatbot configuration that specifies the behavior of a chatbot including the tasks that can be performed by the chatbot, the types of entities relevant to each task, and so on. The conversation may be voice based and use natural language. The conversation engine may load different chatbot configurations to implement different chatbots. The conversation engine receives a conversation engine configuration that specifies the behavior of the conversation engine across chatbots. The system may be a multi-tenant system that allows customization of the chatbots for each tenant.