人工智能数据挖掘在土壤环境科学中的应用基于文献知识图谱的分析

Application of Intelligent Data Mining in Soil Environmental Science: Analysis Based on Literature Knowledge Graph

  • 摘要: 数据的大幅增长和人工智能技术的快速发展为土壤环境科学研究带来新的思路与发展机遇。本文系统梳理人工智能数据挖掘在土壤环境领域的应用研究现状和前沿动态,归纳评述主要研究热点,提出面临的挑战。采用文献计量和知识图谱分析方法对中国知网(CNKI)和Web of Science文献数据库中截至2021年的相关主题文献进行可视化分析。分析结果表明,人工智能数据挖掘在土壤环境领域的应用研究从2000年左右起步,目前正处于快速发展阶段;中国学者在该领域做出大量贡献,成为国际上最重要的研究力量之一。文献知识图谱分析结果显示,土壤污染预测评价、有机碳空间分布预测制图是国内外学者共同关注的热点方向;我国学者在土壤污染溯源、场地土壤污染评价与修复管控决策两个应用方向的研究处于优势地位。随着数据“孤岛”、数据隐私保护、模型可解释性等一系列问题的逐步被克服,基于人工智能技术的数据挖掘将给土壤环境实时监测、评估、预测预警及管理决策带来深远的影响。

     

    Abstract: The massive growth of soil environmental data and rapid development of artificial intelligence technology have brought new ideas and opportunities to soil environmental research. This study reviewed the status and progress of research focused on the application of intelligent data mining (IDM) technologies in the field of soil environmental science. Visualized analysis was conducted on relevant literature up to 2021 from CNKI and Web of Science based on bibliometrics method and knowledge graph. The results showed that studies related to application of IDM technologies in soil environmental field commenced around 2000 and were in the stage of exponential growth. Chinese scholars have made significant contributions in this field as one of the most important research groups in the world. The knowledge graphs of literature indicated that prediction and assessment of soil pollution and mapping of soil organic carbon are two major research topics in the field of soil environmental data mining. Chinese scholars play a leading role in research related to IDM technologies in soil pollution source identification, as well as site soil pollution assessment and remediation decision aids. Although still faced with various challenges such as data isolation, data privacy threatening and model untransparent nature, IDM technologies will bring about profound influence on soil environment monitoring, evaluation, prediction and management in the future.

     

/

返回文章
返回