Integration of Named Entity Extraction Based on Deep Learning for Neo4j Graph Database

The increase in unstructured textual data has created a pressing demand for effective information extraction techniques. This paper explores the integration of Named Entity Extraction (NEE) using deep learning within the Neo4j graph database. Utilizing the Rebel Large Model, we converted raw text into structured knowledge graphs. The primary objective is to evaluate the efficacy of this integration by examining performance metrics, such as process-ing time, graph growth, and entity representation. The findings highlight how the structure and complexity of graphs vary with different text lengths, offering insights into the potential of combin-ing deep learning-based NEE with graph databases for improved data analysis and decision-making.