Automated Information System Dynamic Dialog Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated information systems face challenges in providing accurate and efficient responses due to static conversation dialogs, data duplication, and less accurate answers generated from vast company documents, which lead to inefficient knowledge retrieval and resource expenditure.
Innovation Solution
A computer-implemented method that extracts multiple answers from a knowledge base, detects relationship entities with different values, generates follow-up questions to solicit values, and stores user interactions to refine answers, thereby creating a dynamic knowledge corpus for improved response accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple answers are extracted from a knowledge base to handle complex queries, then answer completeness improves, but data duplication and processing complexity increase
Solution Approach 1:
The patent segments the answer selection process into distinct stages: extracting multiple candidate answers, identifying relationship entities, detecting conditions with different values, and selectively filtering answers based on user-provided values. This segmentation manages complexity by breaking down the challenging task of handling multiple answers into manageable steps.
Solution Approach 2:
The system implements feedback loops where user responses to follow-up questions about relationship entities feed back into the answer selection process. The system continuously refines its answer choices based on user feedback, eliminating answers that don't match the user's specified conditions and presenting progressively more accurate results.
2Adaptability or versatility
If static conversation dialogs are used in automated systems, then system simplicity is maintained, but response accuracy and adaptability decrease
Solution Approach 1:
The patent transforms the static conversation dialog into a dynamic process that adapts to user needs. The system generates follow-up questions based on relationship entities detected in the knowledge base and user preferences, making the dialog flow flexible and responsive rather than following a predetermined static script.
Solution Approach 2:
The system changes parameters of the conversation based on detected conditions. When relationship entities with different values are identified, the system adjusts the dialog by asking follow-up questions about those specific parameters, thereby adapting the conversation to the user's particular situation and needs.
3Measurement precision
If follow-up questions are generated to solicit condition values from users, then answer precision improves, but interaction time increases
Solution Approach 1:
The system applies partial action by generating follow-up questions only for relationship entities that are actually detected as conditions with different values in the knowledge base. It doesn't unnecessarily query all possible parameters, but only those that are relevant to distinguishing between multiple candidate answers, thereby reducing unnecessary interaction time.
4Loss of information
If relationship entities with different values are detected and filtered, then answer relevance improves, but processing complexity increases
Solution Approach 1:
The patent extracts and isolates relationship entities from the knowledge base answers, identifying those that represent conditions with different values. By separating these key discriminating factors from the rest of the answer data, the system can efficiently filter answers based on user input without having to process and compare entire answer texts.
Data Source
AI summary
A method includes: receiving, by a computer device, a user question; extracting, by the computer device and from a knowledge base, multiple answers to the user question; detecting, by the computer device, a first relationship entity that exists in a subset of the multiple answers to the user question; determining, by the computer device, that the first relationship entity is a condition, the condition being a relationship entity that has different values in the subset of the multiple answers; generating, by the computer device, a first follow-up question that solicits a value of the condition; receiving, by the computer device, an answer to the first follow-up question; and determining, by the computer device, a number of the subset of the multiple answers that have a value of the condition that matches the answer to the first follow-up question.