Semi-structured Knowledge Base Query Retrieval via Inverted Index and MRC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intent-driven question answering systems are inefficient for use cases requiring automatic question answering from semi-structured knowledge bases, such as FAQs and help pages, as they necessitate resource-intensive intent recognition and conversation flow graph development.
Innovation Solution
A system that generates answers from semi-structured knowledge bases without intent recognition, allowing developers to provide a root URL or existing question and answer pairs, using a machine reading comprehension module to identify relevant data segments and create an inverted index for efficient query processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If intent-driven question answering systems are used, then the system can handle complex conversational flows, but the resource requirements and system complexity increase significantly
Solution Approach 1:
The patent extracts and removes the intent recognition component from the question answering system. Instead of requiring full intent-driven processing with conversation flow graphs, the system directly matches user questions against pre-stored question-answer pairs from semi-structured knowledge bases, eliminating unnecessary complexity while maintaining effectiveness for FAQ-type queries
Solution Approach 2:
The patent inverts the traditional approach by not starting with intent recognition and then finding answers. Instead, it pre-processes and stores structured question-answer pairs, then directly retrieves matching answers for user queries, reversing the conventional flow to achieve simpler and more efficient operation
2Reliability
If intent recognition and conversation flow graphs are implemented, then the system can process complex queries, but the development resources and time required increase
Solution Approach 1:
The patent applies preliminary action by pre-processing and structuring question-answer pairs from knowledge bases before deployment. The system pre-stores these structured pairs and creates inverted indexes in advance, so that during operation, simple retrieval and matching occur without requiring real-time intent analysis or conversation flow graph construction, significantly reducing development and processing time
3Use of energy by moving object
If semi-structured knowledge bases are used without intent recognition, then resource requirements are reduced, but the system must efficiently organize and retrieve data segments
Solution Approach 1:
The patent introduces an inverted index as an intermediary data structure between the semi-structured knowledge base and the retrieval process. This intermediary organizes data segments by their constituent words or tokens, enabling efficient lookup and retrieval without requiring complex intent recognition, thus reducing resource requirements while maintaining retrieval efficiency
Data Source
AI summary
The disclosed subject matter relates to a system and method for retrieving relevant information in response to a user query without devising intent of the query. The relevant information is contained within a semi-structured database which was populated from Q&A pairs, help web sites, product descriptions and other information from an organizations knowledge base and from which an inverted index is created. The semi-structured data base may be created automatically or entered manually. Upon receiving a user query, data segments are identified (and ranked) via the inverse index and the data segment most similar to the query is provided to a MRC model which reads the segments to determine the portion (span/snippet) of the data segment that addresses the query. This portion is provided to the user in response to the query.