Destination prediction has attracted widespread attention because it can help vehicle-aid systems recommend related services in advance to improve user driving experience.However,the relevant research is mainly based ...Destination prediction has attracted widespread attention because it can help vehicle-aid systems recommend related services in advance to improve user driving experience.However,the relevant research is mainly based on driving trajectory of vehicles to predict the destinations,which is challenging to achieve the early destination prediction.To this end,we propose a model of early destination prediction,DP-BPR,to predict the destinations by users’travel time and locations.There are three challenges to accomplish the model:1)the extremely sparse historical data make it challenge to predict destinations directly from raw historical data;2)the destinations are related to not only departure points but also departure time so that both of them should be taken into consideration in prediction;3)how to learn destination preferences from historical data.To deal with these challenges,we map sparse high-dimensional data to a dense low-dimensional space through embedding learning using deep neural networks.We learn the embeddings not only for users but also for locations and time under the supervision of historical data,and then use Bayesian personalized ranking(BPR)to learn to rank destinations.Experimental results on the Zebra dataset show the effectiveness of DP-BPR.展开更多
In recent years,there have been important developments in the joint analysis of the travel behavior based on discrete choice models as well as in the formulation of increasingly flexible closed-form models belonging t...In recent years,there have been important developments in the joint analysis of the travel behavior based on discrete choice models as well as in the formulation of increasingly flexible closed-form models belonging to the generalized extreme value class.The objective of this work is to describe the simultaneous choice of shopping destination and travel-to-shop mode in downtown area by making use of the cross-nested logit(CNL) structure that allows for potential spatial correlation.The analysis uses data collected in the downtown areas of Maryland-Washington,D.C.region for shopping trips,considering household,individual,land use,and travel-related characteristics.The estimation results show that the dissimilarity parameter in the CNL model is 0.37 and significant at the 95% level,indicating that the alternatives have high spatial correlation for the short shopping distance.The results of analysis reveal detailed significant influences on travel behavior of joint choice shopping destination and travel mode.Moreover,a Monte Carlo simulation for a group of scenarios arising from transportation policies and parking fees in downtown area,was undertaken to examine the impact of a change in car travel cost on the shopping destination and travel mode switching.These findings have important implications for transportation demand management and urban planning.展开更多
An integral connection exists among the mine production planning, the mined material destination, and the ultimate pit limit （UPL） in the mining engineering economy. This relation is reinforced by real information a...An integral connection exists among the mine production planning, the mined material destination, and the ultimate pit limit （UPL） in the mining engineering economy. This relation is reinforced by real information and the benefits it engenders in the mining economy. Hence, it is important to create optimizing algorithms to reduce the errors of economic calculations. In this work, a logical mathematical algorithm that considers the important designing parameters and the mining economy is proposed. This algorithm creates an optimizing repetitive process among different designing constituents and directs them into the maximum amount of the mine economical parameters. This process will produce the highest amount of ores and the highest degree of safety. The modeling produces a new relation between the concept of the cutoff grade, mine designing, and mine planning, and it provides the maximum benefit by calculating the destination of the ores. The proposed algorithm is evaluated in a real case study. The results show that the net present value of the mine production is increased by 3% compared to previous methods of production design and UPL.展开更多
[目的]本研究旨在改善基于深度学习的遥感影像田块语义分割中出现的区域不封闭、边缘不贴合、噪点问题,并进一步修正语义分割的识别错误。[方法]以安徽省阜南县、江苏省淮安市为研究地点,自建了农田田块数据集,引入考虑影像多尺度特征...[目的]本研究旨在改善基于深度学习的遥感影像田块语义分割中出现的区域不封闭、边缘不贴合、噪点问题,并进一步修正语义分割的识别错误。[方法]以安徽省阜南县、江苏省淮安市为研究地点,自建了农田田块数据集,引入考虑影像多尺度特征的尺度分割思想与基于物候学的DESTIN(delineation by fusing spatial and temporal information)分割算法,提出了基于多尺度及DESTIN约束的高分遥感影像农田田块语义分割方法。[结果]多尺度与DESTIN约束下基于深度模型的田块语义分割有效改善模型出现的区域不封闭、边缘不贴合、噪点和块状模糊等问题,一定程度修正了深度模型语义分割的错误识别,IoU指标在2个测试集上分别达到94.08%和90.79%,相较深度模型的遥感影像田块语义分割分别提高1.65%和2.32%,对研究区域的田块提取区域更完整、精度更高。[结论]多尺度及DESTIN约束进一步改善了田块语义分割问题,有助于提高高分遥感影像的田块识别精度。展开更多
基金Project(2018YFF0214706)supported by the National Key Research and Development Program of ChinaProject(cstc2020jcyj-msxmX0690)supported by the Natural Science Foundation of Chongqing,China+1 种基金Project(2020CDJ-LHZZ-039)supported by the Fundamental Research Funds for the Central Universities of Chongqing,ChinaProject(cstc2019jscx-fxydX0012)supported by the Key Research Program of Chongqing Technology Innovation and Application Development,China。
文摘Destination prediction has attracted widespread attention because it can help vehicle-aid systems recommend related services in advance to improve user driving experience.However,the relevant research is mainly based on driving trajectory of vehicles to predict the destinations,which is challenging to achieve the early destination prediction.To this end,we propose a model of early destination prediction,DP-BPR,to predict the destinations by users’travel time and locations.There are three challenges to accomplish the model:1)the extremely sparse historical data make it challenge to predict destinations directly from raw historical data;2)the destinations are related to not only departure points but also departure time so that both of them should be taken into consideration in prediction;3)how to learn destination preferences from historical data.To deal with these challenges,we map sparse high-dimensional data to a dense low-dimensional space through embedding learning using deep neural networks.We learn the embeddings not only for users but also for locations and time under the supervision of historical data,and then use Bayesian personalized ranking(BPR)to learn to rank destinations.Experimental results on the Zebra dataset show the effectiveness of DP-BPR.
基金Projects(JCYJ20120615145601342,JCYJ20130325151523015)supported by Shenzhen Science and Technology Development Funding-Fundamental Research Plan,ChinaProject(2013U-6)supported by Key Laboratory of Eco Planning & Green Building,Ministry of Education(Tsinghua University),China
文摘In recent years,there have been important developments in the joint analysis of the travel behavior based on discrete choice models as well as in the formulation of increasingly flexible closed-form models belonging to the generalized extreme value class.The objective of this work is to describe the simultaneous choice of shopping destination and travel-to-shop mode in downtown area by making use of the cross-nested logit(CNL) structure that allows for potential spatial correlation.The analysis uses data collected in the downtown areas of Maryland-Washington,D.C.region for shopping trips,considering household,individual,land use,and travel-related characteristics.The estimation results show that the dissimilarity parameter in the CNL model is 0.37 and significant at the 95% level,indicating that the alternatives have high spatial correlation for the short shopping distance.The results of analysis reveal detailed significant influences on travel behavior of joint choice shopping destination and travel mode.Moreover,a Monte Carlo simulation for a group of scenarios arising from transportation policies and parking fees in downtown area,was undertaken to examine the impact of a change in car travel cost on the shopping destination and travel mode switching.These findings have important implications for transportation demand management and urban planning.
文摘An integral connection exists among the mine production planning, the mined material destination, and the ultimate pit limit （UPL） in the mining engineering economy. This relation is reinforced by real information and the benefits it engenders in the mining economy. Hence, it is important to create optimizing algorithms to reduce the errors of economic calculations. In this work, a logical mathematical algorithm that considers the important designing parameters and the mining economy is proposed. This algorithm creates an optimizing repetitive process among different designing constituents and directs them into the maximum amount of the mine economical parameters. This process will produce the highest amount of ores and the highest degree of safety. The modeling produces a new relation between the concept of the cutoff grade, mine designing, and mine planning, and it provides the maximum benefit by calculating the destination of the ores. The proposed algorithm is evaluated in a real case study. The results show that the net present value of the mine production is increased by 3% compared to previous methods of production design and UPL.
文摘[目的]本研究旨在改善基于深度学习的遥感影像田块语义分割中出现的区域不封闭、边缘不贴合、噪点问题,并进一步修正语义分割的识别错误。[方法]以安徽省阜南县、江苏省淮安市为研究地点,自建了农田田块数据集,引入考虑影像多尺度特征的尺度分割思想与基于物候学的DESTIN(delineation by fusing spatial and temporal information)分割算法,提出了基于多尺度及DESTIN约束的高分遥感影像农田田块语义分割方法。[结果]多尺度与DESTIN约束下基于深度模型的田块语义分割有效改善模型出现的区域不封闭、边缘不贴合、噪点和块状模糊等问题,一定程度修正了深度模型语义分割的错误识别,IoU指标在2个测试集上分别达到94.08%和90.79%,相较深度模型的遥感影像田块语义分割分别提高1.65%和2.32%,对研究区域的田块提取区域更完整、精度更高。[结论]多尺度及DESTIN约束进一步改善了田块语义分割问题,有助于提高高分遥感影像的田块识别精度。
