Chatbot Authoring via Natural Language Database Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tools for defining chatbots are too low-level and require clients to break down end-user queries into intents, slots, and prompts, making them tedious to use and difficult to scale for data-oriented customer requests.
Innovation Solution
An intuitive method for authoring conversation services that involves receiving a database selection, connecting the chatbot authoring service to a database table, and using natural language queries to translate descriptions of table subjects and columns into SQL statements suitable for querying the database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If clients use existing chatbot authoring tools with intents, slots, and prompts, then they can define basic conversational flows, but the complexity increases and usability decreases for domain experts without programming training
Solution Approach 1:
The patent introduces a natural language query service as an intermediary layer between domain experts and the complex chatbot authoring system. This mediator translates simple natural language descriptions of database tables and columns into structured intents, slots, and prompts automatically, eliminating the need for experts to manually configure complex programming interfaces while maintaining full functionality.
Solution Approach 2:
The system enables self-service authoring by automatically generating chatbot configurations from natural language descriptions. When domain experts provide high-level descriptions of database tables and columns, the natural language query service autonomously breaks down these descriptions into the required intents, slots, and prompts, freeing experts from manual configuration tasks.
2Productivity
If clients break down end-user queries into intents, slots, and prompts manually, then they can control the conversation flow precisely, but the process becomes tedious and time-consuming
Solution Approach 1:
The natural language query service performs preliminary action by automatically analyzing and breaking down natural language descriptions into structured components before the user needs to configure the chatbot. The system pre-processes the high-level descriptions into intents, slots, and prompts, eliminating the need for users to manually perform this time-consuming breakdown process.
Solution Approach 2:
The patent replaces the mechanical manual process of breaking down queries into intents, slots, and prompts with an automated natural language processing system. The mechanical task of manual configuration is substituted by an AI-based service that automatically translates natural language descriptions into the required structured format, dramatically reducing authoring time.
3Measurement precision
If existing tools use text-oriented question answering from imported documents, then they can handle simple queries, but they fail to provide precise answers for data-oriented questions from relational databases
Solution Approach 1:
The patent applies parameter changes by adapting the question answering approach based on the data source type. For relational database tables, the system changes its behavior to automatically identify key columns, understand table structures, and generate precise SQL queries. This allows the same natural language interface to handle both text-oriented documents and data-oriented databases with appropriate precision for each type.
Solution Approach 2:
The natural language query service achieves universality by handling multiple data source types (text documents and relational databases) through a single unified interface. The service adapts its processing logic based on the input type, providing precise answers for both text-oriented questions and data-oriented queries without requiring separate systems for each data type.
4Adaptability or versatility
If specialized chatbot authoring tools are designed for narrow domains like e-Commerce, then they can provide domain-specific functionality, but they cannot work for general domains
Solution Approach 1:
The patent creates a universal chatbot authoring tool that works across multiple domains by using domain-agnostic natural language processing. The system understands database table structures and relationships in a general sense, allowing it to handle various domain-specific tables (customers, products, orders, etc.) without requiring domain-specific configuration or customization, thereby achieving both versatility and ease of use.
Data Source
AI summary
A method of authoring a conversation service for a chatbot and a database includes receiving, from a user, a selection of a database, and connecting an authoring service of the chatbot to a table in the database; outputting, from the authoring service to the user, a question requesting a description of a subject matter of the table; receiving the description of the subject matter of the table; outputting to the user a question requesting an identification of a key column of the table that contains values that represent the subject matter of the table; receiving the identification of the key column of the table; and translating, by a natural language query service, the description of the subject matter of the table and the key column of the table into the conversation service, wherein the conversation service includes SQL statements suitable for querying the database table.