基于知识图谱的番茄种植管理可视化查询  

Visual Query of Tomato Irrigation Based on Knowledge Graph

作　　者：张宇[1,2] 于合龙[1] 郭文忠[2] 林森 文朝武 龙洁花 Zhang Yu;Yu Helong;Guo Wenzhong;Lin Sen;Wen Chaowu;Long Jiehua(College of Information Technology,Jilin Agricultural University,Changchun 130118,China;Intelligent Equipment Technology Research Center,Beijing Academy of Agriculture and Forestry Sciences,Beijing 100097,China)

机构地区：[1]吉林农业大学信息技术学院,长春130118 [2]北京市农林科学院智能装备技术研究中心,北京100097

出　　处：《农机化研究》2024年第3期8-13,共6页Journal of Agricultural Mechanization Research

基　　金：北京市科技计划项目(Z211100004621006);北京市农林科学院青年基金项目(QNJJ202027);宁夏回族自治区重点研发计划项目(2018BBF02024)。

摘　　要：为提高获取番茄种植管理知识的速度与准确率,研究了以图形式描述番茄在不同环境的种植管理,并基于知识图谱构建了番茄种植管理可视化查询系统。该方法利用“自顶向下”和“自底向上”的模块化CREATE解决了Neo4j的缓慢和准确率问题,并利用PyQt框架构建可视化查询界面,通过问题预处理和语义相似度计算输出最合适的番茄种植管理知识。试验结果表明:该方法的平均响应时间和平均准确率比Cypher查询语言分别提高88.33%及1.97%,可操性比Cypher语言友好。研究结果可以在不同环境下为番茄生产管理提供高质量的种植管理建议。In order to improve the speed and accuracy of agricultural workers'acquisition of tomato planting management knowledge,this study described tomato planting management in different environments by graph form.A visual query system for tomato planting management was constructed based on knowledge graph.This method solved the slowness and accuracy problems of Neo4j by using the top-down and bottom-up modular CREATE,and built a visual query interface by using PyQt framework,and outputed the most appropriate tomato planting management knowledge through problem preprocessing and semantic similarity calculation.The experimental results show that the average response time and average accuracy of this method are 88.33%and 1.97%higher than that of Cypher query language,respectively.The operability of this method is more friendly than that of Cypher query language.This study can provide high quality planting management suggestions for tomato production and management in different environments.

关 键 词：知识图谱 Neo4j 相似度计算 问题预处理 可视化查询 

分 类 号：S126[农业科学—农业基础科学] TP319[自动化与计算机技术—计算机软件与理论]