局部空间同位模式在作物种植适宜性调查中的应用
发布时间：2018-11-11 01:12
【摘要】：空间同位模式的局部挖掘算法存在挖掘结果无统计显著性意义、缺少领域知识介入导致挖掘结果针对性不强等问题.通过引入领域专家知识选择候选模式,构建候选模式属性的兴趣度函数作为局部空间统计量,对空间候选模式的兴趣度值聚集区域进行挖掘,进而合并聚集区域的邻域确定模式的空间区域.经过分析及验证,利用该方法对长汀县烟草种植适宜性结果数据集进行处理,发现了主要养分高但不适合种植烟草的异常模式及区域.据此,专家有针对性地提出了加大灌溉基础设施建设、施用石灰和硼肥等改善烟草种植条件的建议.
[Abstract]:The local mining algorithm of spatial colocation pattern has some problems such as no statistical significance of mining results and lack of domain knowledge intervention resulting in weak pertinence of mining results. By introducing domain expert knowledge to select candidate patterns, the interest function of candidate pattern attributes is constructed as local spatial statistics, and the interest value aggregation regions of spatial candidate patterns are mined. Then the spatial region of the neighborhood determination model of the aggregation region is merged. After analysis and verification, the data set of tobacco suitability in Changting County was processed, and the abnormal patterns and regions with high nutrient content but not suitable for tobacco planting were found. Accordingly, the experts put forward some suggestions to improve tobacco planting conditions, such as increasing irrigation infrastructure construction, applying lime and boron fertilizer and so on.
【作者单位】： 福建农林大学资源与环境学院;福建省土壤环境健康与调控重点实验室;
【基金】：福建省科技重大专项(2012NZ0002)
【分类号】：S114
本文编号：2324081
[Abstract]:The local mining algorithm of spatial colocation pattern has some problems such as no statistical significance of mining results and lack of domain knowledge intervention resulting in weak pertinence of mining results. By introducing domain expert knowledge to select candidate patterns, the interest function of candidate pattern attributes is constructed as local spatial statistics, and the interest value aggregation regions of spatial candidate patterns are mined. Then the spatial region of the neighborhood determination model of the aggregation region is merged. After analysis and verification, the data set of tobacco suitability in Changting County was processed, and the abnormal patterns and regions with high nutrient content but not suitable for tobacco planting were found. Accordingly, the experts put forward some suggestions to improve tobacco planting conditions, such as increasing irrigation infrastructure construction, applying lime and boron fertilizer and so on.
【作者单位】： 福建农林大学资源与环境学院;福建省土壤环境健康与调控重点实验室;
【基金】：福建省科技重大专项(2012NZ0002)
【分类号】：S114
【相似文献】
相关重要报纸文章 前1条
1 马海邻;中生代让位模式[N];解放日报;2013年
相关硕士学位论文 前1条
1 谭浩;同位模式空间数据挖掘算法研究及在GIS中的应用[D];中南大学;2011年
，本文编号：2324081
本文链接：https://www.wllwen.com/kejilunwen/nykj/2324081.html
