结合广东省农作物审定品种数据和知识图谱相关技术开展研究。种业是农业产业链的起始环节，是保障国家粮食安全和经济发展的重要支柱，审定品种作为该环节的重要创新资源，经由严格测试和客观评价后予以推广，有效实现种质资源的保护和利用，推动种业高质量发展。随着农业信息化的推进，农业数据量剧增，大数据、人工智能等现代信息技术对提高农业生产效率和优化资源配置带来了突出作用。知识图谱作为人工智能和语义网络的重要分支技术，已广泛应用于各大领域，而农业领域的知识图谱研究，相对侧重作物栽培、水肥管理、病虫害防治等重点问题。本研究基于数据可靠性、实用性、连续性等因素，通过获取广东省农业农村厅公开发布信息，采集了2016—2023年共计8年的广东省农作物审定品种数据作为基础数据，该数据以.doc格式存储，包含大量文字和字符。为便于机器识别及后续知识图谱构建，本研究通过数据清洗去除噪声影响，根据品种特征特性和产量表现提取共性属性，最终整理合并了水稻、玉米、大豆三类农作物审定品种共计823条种质资源数据，并以.xlsx和.json两种格式存储为结构化数据。为验证数据有效性，本研究采用Neo4j图形数据库成功构建了广东省主要农作物审定品种知识图谱。相关科研和生产单位可基于本数据集建立农作物审定品种专家知识库，并通过数据库扩充、多源数据融合等操作，构建面向具体农业任务的智能问答、管理决策、信息推荐等智慧服务。

数据摘要：

This study is carried out in combination with the data of crops approved varieties in Guangdong Province and related technologies of knowledge map. Seed industry is the initial link of agricultural industrial chain and an important pillar to ensure national food security and economic development. As an important innovative resource in this link, approved varieties are popularized after strict testing and objective evaluation, which effectively realizes the protection and utilization of germplasm resources and promotes the high-quality development of seed industry. With the advancement of agricultural informatization, the amount of agricultural data has increased dramatically, and modern information technologies such as big data and artificial intelligence have played a prominent role in improving agricultural production efficiency and optimizing resource allocation. As an important branch technology of artificial intelligence and semantic network, knowledge mapping has been widely used in various fields, while the research of knowledge mapping in agricultural field focuses on key issues such as crop cultivation, water and fertilizer management, pest control and so on. Based on the reliability, practicability, continuity and other factors of data, this study collected the eight-year crop variety data of Guangdong Province from 2016 to 2023 as basic data by obtaining the information publicly released by the Guangdong Provincial Department of Agriculture and Rural Affairs. The data was stored in. doc format and contained a lot of characters and characters. In order to facilitate machine identification and subsequent knowledge map construction, this study removed the influence of noise by data cleaning, and extracted common attributes according to the characteristics and yield performance of varieties. Finally, 823 germplasm resources data of three crops approved varieties by rice, corn and soybean were sorted and merged, and stored as structured data in. xlsx and. json formats. In order to verify the validity of the data, the knowledge map of main crops approved varieties in Guangdong Province was successfully constructed by using the graphic database: Neo4j. Relevant scientific research and production units can establish an expert knowledge base of crops approved varieties based on this data set, and build intelligent services such as intelligent question and answer, management decision and information recommendation for specific agricultural tasks through database expansion and multi-source data fusion.

Data summary: