摘要
研究基于重庆垫江南部地区3250份土壤表层样品的54类地球化学数据,应用Python编程语言进行Apriori关联规则算法研究,尝试通过大数据分析的方法提取出隐藏在数据集中的土壤养分元素与其他各元素之间的组合规律,为垫江地区的土地利用规划、生态环境保护、现代农业发展以及土地污染防治等工作的开展提供科学依据。通过Apriori方法进行的重庆垫江南部土壤元素数据分析最终反映出各元素之间共存在八种关联关系。Fe、Zn、Co、V、Mg、Cu六类土壤养分元素的关联性极强,会出现共同富集的现象;土壤中Mo、Ca元素的缺乏与Li、I、U、P元素的匮乏存在一定关联;其他土壤养分元素则相对独立,无法建立与其他元素之间的关联规则。相较于传统地球化学分析方法,关联规则算法具有快捷、准确、完全基于数据深入挖掘等特点,对于未来土壤养分元素多寡的指示以及富集等级的划分具有重要的学术与应用价值。
Based on 54 types of geochemical data from 3250 topsoil samples in South Dianjiang County of Chongqing,Apriori association rule algorithm is carried out using Python programm language to extract the combination rule between the soil nutrient elements hidden in the data set and other elements,so as to provide a scientific basis for the implementation of land use planning,ecological environmental protection,modern agricultural development and land contamination prevention and controlling in the Dianjiang area.The analysis of soil elemental data in South Dianjiang of Chongqing,conducted by the Apriori method ultimately reflects a total of eight correlations among the elements.The research result shows that six types of soil nutrient elements,Fe,Zn,Co,V,Mg and Cu,are strongly correlated and might see synchronous enrichment,and that the lack of Mo and Ca elements in the soil is related to the deficiency in Li,I,U and P elements.It is also impossible to establish the association rule between other relatively independent soil nutrient elements and other elements.Compared to traditional geochemical analysis methods,the association rule algorithm is fast,accurate,and based entirely on deep data mining.There is an important scientific and practical value for the indication of the content of soil nutrient elements and classification of enrichment levels in the future.
作者
侯典吉
雷冲
王凯伟
蒙丽
HOU Dianji;LEI Chong;WANG Kaiwei;MENG Li(Qingdao Sparta Analysis&Test Co.,Ltd,Qingdao 266000,China;Nanjiang Hydro-geology and Engineering Geology Team,Chongqing Bureau of Geology and Mineral Exploration and Development,Chongqing 401121,China;Key Laboratory of Submarine Geoscience and Prospecting Technique,MOE and College of Marine Geosciences,Ocean University of China,Qingdao 266100,China;Chongqing planning and Natural Resources Bureau,Chongqing 401147,China)
出处
《矿产与地质》
2024年第3期594-601,共8页
Mineral Resources and Geology
基金
重庆市2020年度第一批地质矿产勘查服务项目(土地质量地质调查类)(编号:20C00392)资助。
