Year: 2025
Type: conference-paper
Abstract: This paper introduces "GraphQuery," an innovative system that integrates natural language processing capabilities of Large Language Models (LLMs) with the robust architecture of Neo4j graph databases and the visual representation power of Python libraries including NetworkX and Matplotlib.GraphQuery transforms conversational user inputs into executable Cypher queries, facilitating seamless interaction with graph structures without requiring specialized query language knowledge.The system renders dynamic visualizations based on databaseresponses, allowing intuitive exploration of complex relational networks.GraphQuery excels at processing multi-levelrelationships and identifying subtle patterns within densely connected data.When evaluated on structured clinical documentation, the system demonstrated superior capability in extracting comprehensive patient information and potential diagnoses compared to conventional natural language processing approaches.The platform serves as a versatile tool for educational purposes, data analysis, and research support by simplifying access t o sophisticated graph datasets.
Authors: Shubham Sharma, Geerija Lavania, Vibhor Bhatt, Vaibhav Goyal
Language: English
Topic: Graph Theory and Algorithms
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4412362152