Semantic Graph Database for Chatbot Response Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for natural language processing in chatbots are inefficient due to the large amount of data required for storing and processing entire conversation logs, making real-time communication challenging.
Innovation Solution
A method utilizing a semantic graph database is employed, where past user utterances are transformed into dependency parse trees, integrated into vertices, and relationships are stored as edges, enabling quick retrieval and generation of chatbot responses through natural language generation algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire conversation log is stored and processed, then the chatbot can provide comprehensive responses, but the data processing time increases and real-time performance deteriorates
Solution Approach 1:
The patent segments the conversation log into individual utterances and stores them as separate entities in the database. Each utterance is processed and stored independently with its semantic structure, allowing the system to retrieve only relevant segments rather than processing the entire conversation log, thus reducing processing time while maintaining response comprehensiveness.
Solution Approach 2:
The patent performs preliminary processing of conversation logs by generating dependency parse trees and extracting semantic structures in advance. This preprocessing transforms raw text into structured data that can be quickly queried and matched during real-time interactions, eliminating the need for complex processing during actual chatbot responses.
2Loss of information
If the entire conversation log is stored, then no information is lost, but the data volume becomes excessively large
Solution Approach 1:
The patent extracts essential semantic information from conversation logs by generating dependency parse trees and identifying key linguistic structures. Only the extracted semantic features and structural relationships are stored in the database, rather than storing the complete raw text, which significantly reduces storage requirements while preserving the essential information needed for meaningful responses.
Solution Approach 2:
The patent creates structured representations (dependency parse trees) that copy and represent the essential semantic structure of conversation utterances. These structured copies capture the meaningful relationships between words and phrases without requiring storage of the original verbose text, enabling efficient storage and retrieval of conversation information.
3Device complexity
If simple text storage is used, then the system is simple to implement, but the natural language processing capability is insufficient
Solution Approach 1:
The patent introduces dependency parse trees as an intermediary structure between raw text input and chatbot response generation. These parse trees serve as a mediator that captures the semantic relationships and structural patterns in natural language, enabling the system to understand and process user intent more effectively without requiring overly complex processing architectures.
Data Source
AI summary
Disclosed is a method for generating an utterance of a chatbot based on a semantic graph database and executed by a computer, includes generating the semantic graph database by using a past utterance of a user, and generating the utterance of the chatbot for responding to a new utterance of the user, by using the semantic graph database.