Knowledge Graph Question Answering With Visual Reasoning Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing question answering solutions suffer from low accuracy and flexibility, poor visualization in the interaction process, and inadequate interaction effects, particularly in human-computer interactions, with users unable to intuitively follow the reasoning process and receive relevant recommendations.
Innovation Solution
A human-computer interaction method that utilizes a knowledge graph to display a sub-graph of entities and relationships, incorporating directional identifiers to illustrate the reasoning path, and recommends objects based on user attributes and behaviors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing question answering solutions are used, then the system can provide automated answers, but the accuracy and flexibility are low
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary between the user's question and the answer generation process. The knowledge graph stores structured entities and relationships, allowing the system to perform accurate reasoning based on pre-established knowledge connections. This intermediary structure enables both high accuracy through structured reasoning and flexibility through configurable query paths.
Solution Approach 2:
The patent replaces traditional text-based question answering mechanisms with a graph-based knowledge representation system. Instead of relying on natural language processing alone, the system uses structured entity-relationship graphs with directional identifiers to represent reasoning paths, substituting mechanical text analysis with structured graph traversal and reasoning.
2Loss of information
If traditional question answering systems are used, then answers can be provided, but visualization in the interaction process is poor
Solution Approach 1:
The patent uses color-coded directional identifiers in the knowledge graph visualization to represent different reasoning paths and entity types. Different colors help users distinguish between various types of relationships and reasoning steps, making the interaction process visually intuitive and easier to understand.
Solution Approach 2:
The patent transforms the abstract reasoning process into a visual two-dimensional graph representation. By displaying entities as nodes and relationships as edges with directional identifiers, the system adds a visual dimension to the reasoning process, allowing users to intuitively follow the logical flow from question to answer.
3Quantity of substance
If comprehensive knowledge graphs are displayed, then complete information is provided, but screen utilization and display effects deteriorate
Solution Approach 1:
The patent extracts and displays only the relevant sub-graph portion of the knowledge graph that pertains to the current question and reasoning process. Instead of displaying the entire knowledge graph, the system identifies and extracts the minimal necessary subset of entities and relationships, maintaining information completeness for the current context while optimizing screen utilization.
Solution Approach 2:
The patent segments the large knowledge graph into manageable sub-graphs based on the current query context. By dividing the comprehensive knowledge base into relevant fragments and displaying only the necessary segments, the system maintains complete information for the current reasoning task while improving display effectiveness and screen utilization.
Data Source
AI summary
The present disclosure provides a human-computer interaction method and apparatus, computer-readable storage medium and electronic device, and relates to the technical field of computer question answering. The human-computer interaction method includes receiving an input question; extracting entities and relationships involved in the input question; querying an answer to the input question in a knowledge graph according to the entities and the relationships; and displaying a sub-graph of the knowledge graph. The sub-graph of the knowledge graph includes entities involved in the input question and the answer, and a directional identifier, wherein the directional identifier is configured to identify a reasoning path corresponding to the query process.