Taking Fengdu County in Three-Gorge Reservoir region as an example,this paper applies the dynamic clustering analysis model, and divides land resource utilization of Fengdu County into three zones: Yangtze River valle...Taking Fengdu County in Three-Gorge Reservoir region as an example,this paper applies the dynamic clustering analysis model, and divides land resource utilization of Fengdu County into three zones: Yangtze River valley economic zone,low mountains and hills agricultural zone,and low and middle agricultural,forestry and animal husbandry zone. It also analyzes current situation and existing problems in the land utilization of every zone,and finally puts forward corresponding recommendations.展开更多
针对旅客公交下车站点数据缺失问题,提出融合公交与地铁刷卡数据的旅客公交出行链三级补全策略。该补全策略包括换乘行程链补全策略、公交出行链补全预测模型和最大概率下车预测模型共三个级别。利用同日及隔日换乘场景下的空间位置约...针对旅客公交下车站点数据缺失问题,提出融合公交与地铁刷卡数据的旅客公交出行链三级补全策略。该补全策略包括换乘行程链补全策略、公交出行链补全预测模型和最大概率下车预测模型共三个级别。利用同日及隔日换乘场景下的空间位置约束补全常规公交下车点;引入改进基于密度的聚类算法DBSCAN(Density-Based Spatial Clustering of Applications with Noise),结合加权几何中心构建公交下车站点优化模型;基于公交线路下车概率分布构建最大概率下车模型,通过该三级补全策略实现公交下车站点全流程补全体系的搭建。基于真实出行数据进行试验,结果表明,所提的补全策略可使公交出行链补全准确率达到92.5%,较传统算法准确率提高10%以上,且超80%的误差站点集中于实际站点±1站范围内,证明了其在准确率和鲁棒性方面的优越性。展开更多
文摘Taking Fengdu County in Three-Gorge Reservoir region as an example,this paper applies the dynamic clustering analysis model, and divides land resource utilization of Fengdu County into three zones: Yangtze River valley economic zone,low mountains and hills agricultural zone,and low and middle agricultural,forestry and animal husbandry zone. It also analyzes current situation and existing problems in the land utilization of every zone,and finally puts forward corresponding recommendations.
文摘针对旅客公交下车站点数据缺失问题,提出融合公交与地铁刷卡数据的旅客公交出行链三级补全策略。该补全策略包括换乘行程链补全策略、公交出行链补全预测模型和最大概率下车预测模型共三个级别。利用同日及隔日换乘场景下的空间位置约束补全常规公交下车点;引入改进基于密度的聚类算法DBSCAN(Density-Based Spatial Clustering of Applications with Noise),结合加权几何中心构建公交下车站点优化模型;基于公交线路下车概率分布构建最大概率下车模型,通过该三级补全策略实现公交下车站点全流程补全体系的搭建。基于真实出行数据进行试验,结果表明,所提的补全策略可使公交出行链补全准确率达到92.5%,较传统算法准确率提高10%以上,且超80%的误差站点集中于实际站点±1站范围内,证明了其在准确率和鲁棒性方面的优越性。
