| 39 | 0 | 39 |
| 下载次数 | 被引频次 | 阅读次数 |
随着乡村振兴战略的深入推进，特色种植产业已成为推动农村经济发展的重要抓手。然而，在实践推广过程中普遍面临种植知识获取渠道有限与农业专家资源匮乏的双重困境，亟须一种高效且准确的种植作物推荐方法，实现适合村庄特色的种植作物快速有效推荐。本文提出结合知识图谱与关系图卷积网络的特色种植作物推荐模型。首先，提取包含基础条件、地形特征、土壤特征、气候特征4个维度的22项指标，构建本体结构实现知识的形式化表达；其次，依托“一村一品”示范村镇数据构建知识图谱，涵盖1 915个26种类型的实体节点与37 002条25类关联关系，并采用关系图卷积网络与孪生网络联合训练学习村庄和种植作物的特征向量；最后，通过链接预测任务建模村庄特征与作物的适配关系，实现候选作物的推荐概率排序。实验结果表明，本文模型Top-3推荐准确率达95%,可为特定村庄的振兴发展提供实际指导，具有实践应用价值。
Abstract：With the in-depth implementation of the rural revitalization strategy, specialty crop cultivation has become a crucial driver for rural economic development.However, the practical promotion of such crops is often hindered by the dual challenges of limited access to cultivation knowledge and a scarcity of agricultural experts.Thus, there is an urgent need for an efficient and accurate crop recommendation method to support the rapid identification of suitable crops tailored to village-specific conditions.In this paper, a characteristic crop recommendation model integrating knowledge graph and relational graph convolutional networks is proposed.First, we establish an evaluation system consisting of 22 indicators across four dimensions(including basic conditions, topographical features, soil characteristics, and climate environment),and construct ontological structures to enable formal knowledge representation.Next, using data from "One Village, One Product" demonstration villages, we build a knowledge graph comprising 1 915 entity nodes of 26 types and 37 002 relationships across 25 categories.The model employs relational graph convolutional networks jointly trained with siamese networks to learn feature vectors for villages and crops.Finally, we model the adaptation relationship between village characteristics and crops as a link prediction task, enabling probabilistic ranking of candidate crops.Experimental results show that the proposed model achieves a Top-3 recommendation accuracy of 95%,demonstrating its practical value in providing actionable guidance for village-specific revitalization.
[1] 朱益军.基于案例推理的乡镇产业融合经验挖掘方法与辅助决策系统研究[D].重庆：重庆大学，2021.
[2] 龙思成.乡镇政府推进乡村振兴示范村建设研究：以兴安县J镇为例[D].桂林：广西师范大学，2023.
[3] 龙花楼.《农业专业村的地域经济效应研究：以河南省种植业专业村为例》评介[J].经济地理，2023,43(8):239.
[4] 谢臻，张凤荣，陈松林，等.中国乡村振兴要素识别与发展类型诊断：基于99个美丽乡村示范村的信息挖掘分析[J].资源科学，2019,41(6):1048-1058.
[5] 陈实秋.基于推荐算法的种植推荐系统研究[D].武汉：武汉轻工大学，2023.
[6] Bhat S A,Hussain I,Huang Nenfu.Soil suitability classification for crop selection in precision agriculture using GBRT-based hybrid DNN surrogate models[J].Ecological Informatics,2023,75:102109.DOI:10.1016/j.ecoinf.2023.102109.
[7] 唐嘉平，刘钊.基于GIS的特色经济作物种植适宜性评价系统[J].农业系统科学与综合研究，2002,18(1):9-12.
[8] Chen Wei,Li Zhemin.Agent-based simulation for agricultural learning resource recommendation based on geographical similarities[C]//Emerging Technologies for Education (SETE 2016).Cham:Springer,2017:558-563.
[9] Pudumalar S,Ramanujam E,Rajashree R H,et al.Crop recommendation system for precision agriculture[C]//2016 Eighth International Conference on Advanced Computing (ICoAC).Chennai:IEEE,2017:32-36.
[10] Motwani A,Patil P,Nagaria V,et al.Soil analysis and crop recommendation using machine learning[C]//2022 International Conference for Advancement in Technology (ICONAT).Goa:IEEE,2022:1-7.
[11] Kiruthika S,Karthika D.IOT-BASED professional crop recommendation system using a weight-based long-term memory approach[J].Measurement:Sensors,2023,27:100722.DOI:10.1016/j.measen.2023.100722.
[12] Zhang Qiusi,Li Bo,Zhang Yong.Suitability evaluation of crop variety via graph neural network[J].Computational Intelligence and Neuroscience,2022,2022:5614974.
[13] 向延平.我国“一村一品”地理分布与数量特征分析：兼论区域内生发展[J].地域研究与开发，2014,33(2):51-53.
[14] 王铁，邰鹏飞.山东省国家级乡村旅游地空间分异特征及影响因素[J].经济地理，2016,36(11):161-168.
[15] 丁华，梁婷，薛艳青，等.基于ArcGIS的陕西省乡村旅游空间分布与发展特色研究：以231个省级乡村旅游示范村为例[J].西北师范大学学报(自然科学版),2020,56(3):110-117.
[16] 李玉萍，叶露，梁伟红，等.我国茄果类蔬菜“一村一品”发展现状与路径[J].中国蔬菜，2023(4):1-7.
[17] 朱乾坤，乔家君，韩冬，等.河南省专业村空间格局与集聚测度[J].地理科学进展，2022,41(3):396-408.
[18] 杨坤，芮旸，李宜峰，等.基于共生理论的中国特色保护类村庄振兴类型细分研究[J].地理科学进展，2021,40(11):1861-1875.
[19] Schlichtkrull M,Kipf T N,Bloem P,et al.Modeling relational data with graph convolutional networks[C]//The Semantic Web (ESWC 2018).Cham:Springer,2018:593-607.
[20] Bromley J,Guyon T,Lecun Y,et al.Signature verification using a "Siamese" time delay neural network[C]//Proceedings of the 7th International Conference on Neural Information Processing Systems.San Francisco:Morgan Kaufmann Publishers Inc,1993:737-744.
[21] 王昊奋，漆桂林，陈华钧.知识图谱：方法、实践与应用[M].北京：电子工业出版社，2019.
[22] 周爽，吴娜琳.中国特色种植村镇的空间分布及其成因：以国家级示范性专业村镇为例[J].经济地理，2021,41(4):137-147.
[23] 曹智，刘彦随，李裕瑞，等.中国专业村镇空间格局及其影响因素[J].地理学报，2020,75(8):1647-1666.
[24] 张哲晰，穆月英，赵霞.蔬菜专业村形成及竞争优势发挥的逻辑框架：基于产业集聚视角的实证[J].中国农业资源与区划，2020,41(9):220-229.
[25] 赵永琪，梁俊芬.广东省荔枝专业村空间格局及影响因素研究[J].中国农学通报，2023,39(20):155-164.
[26] 乔家君，刘洋，朱乾坤，等.基于文本分析的河南省专业村发展影响因素研究[J].河南大学学报(自然科学版),2022,52(5):505-517.
[27] 周灿，李小建.河南省农业专业村发展与地理环境关系研究[J].地域研究与开发，2015,34(4):130-135.
[28] Yang Ren,Zhang Xin,Xu Qian.Spatial distribution characteristics and influencing factors of agricultural specialized villages in Guangdong province,China[J].Chinese Geographical Science,2022,32(6):1013-1034.
[29] 李小建，罗庆，樊新生.农区专业村的形成与演化机理研究[J].中国软科学，2009(2):71-80.
[30] Yang Zhe,Chu Lin,Wang Chen,et al.What drives the spatial heterogeneity of cropping patterns in the Northeast China:the natural environment,the agricultural economy,or policy?[J].Science of the Total Environment,2023,905:167810.DOI:10.1016/j.scitotenv.2023.167810.
[31] 吴娜琳，乔家君，梁展鹏，等.河南省不同演化类型专业村的特征及其影响因素分析[J].地理研究，2023,42(6):1560-1576.
[32] Kumar R,Singh M P,Kumar P,et al.Crop selection method to maximize crop yield rate using machine learning technique[C]//2015 International Conference on Smart Technologies and Management for Computing,Communication,Controls,Energy and Materials (ICSTM).Avadi:IEEE,2015:138-145.
[33] 吴志文.药材种植推荐系统的研究与实现[D].北京：北京林业大学，2020.
[34] 迈克尔·波特.国家竞争优势[M].李明轩，邱如美，译.北京：中信出版社，2007.
[35] Salton G,Wong A,Yang C S,et al.A vector space model for automatic indexing[J].Communications of the ACM,1975,18(11):613-620.
[36] 卫操.基于图中图神经网络的软件系统中类交互关系预测研究[D].武汉：湖北大学，2023.
[37] Wang Xiao,Ji Houye,Shi Chuan,et al.Heterogeneous graph attention network[C]//The World Wide Web Conference.New York:Association for Computing Machinery,2019:2022-2032.
[38] Hu Ziniu,Dong Yuxiao,Wang Kuansan,et al.Heterogeneous graph transformer[C]//Proceedings of The Web Conference 2020.New York:Association for Computing Machinery,2020:2704-2710.
[39] Li Hao,Yue Peng,Wu Haoru,et al.A question-answering framework for geospatial data retrieval enhanced by a knowledge graph and large language models[J].International Journal of Digital Earth,2025,18(1):2510566.DOI:10.1080/17538947.2025.2510566.
[40] 顾军华，王锐，李宁宁，等.融合协同过滤信息的知识图注意力网络[J].计算机应用，2022,42(4):1087-1092.
[41] 师博雅，梁光成，孙宇健，等.基于协同知识嵌入注意网络的推荐算法研究[J].计算机应用与软件，2024,41(4):297-305.
[42] Wang Qiang,Yu Zhihang,Wang Shu,et al.Unveiling optimal SDG pathways:an innovative automated recommendation approach integrating graph pruning,intent graph,and attention mechanism[J].International Journal of Digital Earth,2025,18(1):2513048.DOI:10.1080/17538947.2025.2513048.
基本信息:
中图分类号:TP391.1;TP183;P208;S126
引用信息:
[1]谭永滨,杨凯,胡囡,等.知识图谱与RGCN支持的村庄特色种植作物推荐[J].地理与地理信息科学,2026,42(03):19-26.
基金信息:
国家自然科学基金项目(42361067); 东华理工大学2024年度研究生创新专项(YC2024-S486); 东华理工大学2025年度研究生创新专项(DHYC-2025100); 福建省属公益类科研院所专项(2023R1030006)
2026-06-02
2026-06-02
2026-06-02