Term-cluster knowledge graph for chatbot query resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing knowledge graph systems for chatbots struggle to address ambiguous user queries effectively, as they rely on predefined relationships and labeled data, which are not always available, and lack the ability to automatically generate augmented knowledge graphs with term clusters that provide additional context for conversation guidance.
Innovation Solution
A computer-implemented method for generating an augmented knowledge graph by extracting second terms from a domain corpus that do not meet predetermined relationships, forming term clusters, and linking them to nodes in the knowledge graph, enabling chatbots to address both non-ambiguous and ambiguous queries by ranking and presenting relevant terms to users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If knowledge graph systems rely on predefined relationships and labeled data, then the structure and organization of information is maintained, but the ability to address ambiguous user queries deteriorates
Solution Approach 1:
The system segments terms into two categories: first terms that form the core knowledge graph nodes with predefined relationships, and second terms that form term clusters associated with nodes. This segmentation allows the system to maintain structured relationships while adding flexible contextual terms to handle ambiguous queries.
Solution Approach 2:
The system adds a new dimension to the traditional knowledge graph by introducing term clusters as an additional layer. Instead of only having nodes with predefined relationships, the system creates associated term clusters that provide alternative contextual dimensions, enabling the graph to address ambiguous queries from multiple angles.
2Adaptability or versatility
If term clusters are extracted from domain corpus without predefined relationships, then additional context for ambiguous queries is provided, but the complexity of knowledge graph construction increases
Solution Approach 1:
The system automatically extracts second terms from the domain corpus and forms term clusters without requiring manual labeling or predefined relationships. The extraction process is self-service, using computational methods to identify and organize terms that associate with knowledge graph nodes, thereby reducing manual intervention while providing rich contextual information.
Solution Approach 2:
The system replaces manual knowledge graph construction and term association with automated computational extraction. Instead of mechanically organizing terms through predefined relationships, the system uses algorithms to extract second terms from the domain corpus and automatically form term clusters, reducing construction complexity while maintaining adaptability.
3Measurement precision
If semantic features are automatically identified and linked to term pairs, then the accuracy of query resolution is improved, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary extraction and identification of semantic features during the knowledge graph construction phase. By pre-identifying and linking semantic features to term pairs before actual query processing, the system reduces computational burden during query resolution, thereby improving accuracy while minimizing processing time loss.
Solution Approach 2:
The system focuses semantic feature identification on critical term pairs that are most relevant to query resolution rather than processing all possible term pairs exhaustively. This partial action approach maintains high accuracy for important queries while reducing overall processing time and computational resource consumption.
Data Source
AI summary
Given a skeleton knowledge graph including first terms in respective nodes, wherein first terms in connected nodes have a predetermined relationship, an augmented knowledge graph is formed by a first computing device by extracting a second term from a domain corpus to form a term cluster linked with a respective node of the knowledge graph. The second term is associated with the first term of a respective node in the domain corpus while not meeting the predetermined relationship. A semantic feature between the second term and the associated first term is identified in the domain corpus and linked to the pair of the second term and the first term in the augmented knowledge graph. The augmented knowledge graph is useable by a second computing device, which may or may not be the same as the first computing device, to drive a conversation between a chatbot and user.