Natural Language Model Mapping Queries to Knowledge Graph Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems fail to effectively allow users to search for technical information in software applications using their own words, as they lack compatibility between chatbot systems and knowledge graphs, leading to inadequate responses to long-tail queries, which often require live support and increase user wait times and costs for software providers.
Innovation Solution
A system that combines a natural language model and a response orchestrator to map user queries to nodes in a knowledge graph, enabling automated responses by training the natural language model using knowledge graph data and integrating it with chatbots to handle long-tail queries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a knowledge graph is used to store technical information, then information organization and retrieval are improved, but the system cannot process natural language queries from users
Solution Approach 1:
A natural language processing module is introduced as an intermediary between users and the knowledge graph. This module converts natural language queries into structured search queries that the knowledge graph can process, enabling users to query technical information in their own words while maintaining the organized structure of the knowledge graph.
Solution Approach 2:
The system integrates multiple functions into a unified architecture: the knowledge graph stores and organizes technical information, the natural language processing module interprets user queries, and the chatbot interface provides conversational responses. This multi-functional system handles both structured information retrieval and unstructured natural language communication.
2Ease of operation
If a chatbot system is implemented to handle user queries, then user service is improved, but the chatbot cannot effectively respond to long-tail queries without live support
Solution Approach 1:
The chatbot system is merged with the knowledge graph and natural language processing module. The chatbot handles conversational interaction while the knowledge graph provides accurate technical information and the NLP module bridges natural language queries with structured data, enabling the system to handle long-tail queries accurately without live support.
Solution Approach 2:
The system enables automated self-service for handling user queries. The integrated architecture allows the chatbot to independently process natural language queries, search the knowledge graph for relevant technical information, and provide accurate responses without requiring live agent intervention, even for complex long-tail queries.
3Reliability
If live support is provided for complex queries, then query resolution accuracy is improved, but user wait time and support costs increase
Solution Approach 1:
The automated system handles complex queries through self-service capabilities. The knowledge graph contains comprehensive technical information that can resolve complex queries autonomously, and the integrated NLP module interprets complex natural language queries accurately, eliminating the need for live agent intervention and reducing user wait time while maintaining resolution accuracy.
Solution Approach 2:
The knowledge graph is pre-populated with comprehensive technical information organized in a structured manner. This preliminary organization of information enables the system to quickly retrieve accurate responses to complex queries without requiring live agent analysis, reducing user wait time while maintaining high resolution accuracy.
Data Source
AI summary
Certain aspects of the present disclosure provide techniques for mapping natural language to stored information. The method generally includes receiving a long-tail query comprising a natural language utterance from a user of an application associated with a set of topics and providing the natural language utterance to a natural language model configured to identify nodes of a knowledge graph. The method further includes, based on output of the natural language model, identifying a node of a knowledge graph associated with the natural language utterance, wherein the output of the natural language model includes a node identifier for the node of the knowledge graph and providing the node identifier to the knowledge engine. The method further includes receiving a response associated with the node of the knowledge graph from the knowledge engine and transmitting the response to the user in response to the long-tail query.