基于元数据与知识图谱的省级气象数据管理方法及应用

Leveraging Metadata and Knowledge Graphs for Provincial Meteorological Data Management: Methods and Applications

摘要: 针对省级气象业务中数据来源多样、系统分散、语义关联不足以及数据服务智能化水平不高等问题，提出了一种基于元数据与知识图谱融合的省级气象数据知识化管理与智能应用方法。该方法以气象领域本体为统一语义模型，通过对多源异构气象元数据的语义映射与融合，构建省级气象知识图谱，实现气象数据从“以存储管理为主”向“以知识组织与关联服务为主”的转变。在此基础上，设计并实现了语义检索、关联发现和智能问答等应用场景。验证结果表明，该方法能够有效打破省级气象业务系统间的数据壁垒，显著提升复杂业务查询与信息获取效率，可为省级气象数据资源的精细化管理和智能服务提供了可行技术路径参考。

Abstract: Addressing issues in provincial-level meteorological operations such as diverse data sources, fragmented systems, insufficient semantic associations, and a lack of intelligent data services, this paper proposes a method for knowledge-based management and intelligent application of provincial meteorological data, integrating metadata with knowledge graphs. Using a meteorological domain ontology as a unified semantic model, the method semantically maps and fuses multi-source heterogeneous meteorological metadata to construct a provincial meteorological knowledge graph. This shifts the focus of meteorological data from “storage-centric management” to “knowledge organization and association services.” On this basis, application scenarios including semantic search, association discovery, and intelligent Q&A are designed and implemented. Application verification results demonstrate that the method can effectively break down data silos between provincial meteorological business systems, significantly improve the efficiency of complex operational queries and information retrieval, and provide a feasible technical pathway for the refined management and intelligent services of provincial meteorological data resources.