Tai’an City is rich in tourism resources,and there are many A-class tourist attractions.Based on the functions and characteristics of geographic information system(GIS),as well as the spatial distribution characteris...Tai’an City is rich in tourism resources,and there are many A-class tourist attractions.Based on the functions and characteristics of geographic information system(GIS),as well as the spatial distribution characteristics and distribution structure of tourist attractions,the research methods of nearest-neighbor distance,connectivity,accessibility and closeness analysis are used to analyze road traffi c distribution between tourist attractions in Tai’an City.①The R value of the nearest-neighbor point is 0.27,which proves that tourist attractions in Tai’an City are in the state of clustering,and the spatial distance of tourist attractions in clustering area is small.②Theβindex andγindex of connectivity are 1.42 and 0.58,respectively,indicating that the traffi c network connectivity between tourist attractions is low,there is no dense traffi c network,and the optimal combination of tourism resources in tourist attractions is not high.③The accessibility index Ai is 37.68 km,indicating that the accessibility is low and the traffi c between tourist attractions is not perfect.The research results will provide the reference for scientifi c,systematic and objective comprehensive evaluation of existing A-class scenic areas and traffi c conditions in Tai’an City.展开更多
The performance level of traffic control systems highly relies on accurate short-term network-wide traffic prediction.Signal processing techniques have been widely integrated with deep learning algorithms for this pur...The performance level of traffic control systems highly relies on accurate short-term network-wide traffic prediction.Signal processing techniques have been widely integrated with deep learning algorithms for this purpose,but no study has focused on how the parameter of wavelet decomposition order(level)affects the prediction robustness given the training data limitations.This study proposes a hybrid framework for short-term network-wide traffic state prediction,which applies multi-layer perceptron(MLP)neural networks to implicitly capture the traffic network co-movement patterns.Then,the framework is complemented by a seasonal auto-regressive integrated moving average(SARIMA)model,extracting the location-specific features,including the local seasonality and stochastic disturbances.Besides,the hybrid framework is used to explore the association between the traffic prediction accuracy and wavelet decomposition order using an order-adaptive discrete haar wavelet transform(DHWT).The proposed method was validated over four open-access datasets with different training data characteristics in Paris and Madrid urban areas.The results indicated that the hybrid framework significantly improved the predictive accuracy of the benchmark deep learning algorithms.The forecasts for the low-resolution dataset experienced a noticeable improvement once higher wavelet orders were applied.The statistical analysis revealed the moderating effects of the training sample size and data spatial resolution on the link between the wavelet decomposition orders and the model predictive performance.Nevertheless,a combination of pre-processing order of two,and post-processing order of three led to satisfactory results in the cases where sufficient historical data were available.展开更多
Information centric networking(ICN) is a new network architecture that is centred on accessing content. It aims to solve some of the problems associated with IP networks, increasing content distribution capability and...Information centric networking(ICN) is a new network architecture that is centred on accessing content. It aims to solve some of the problems associated with IP networks, increasing content distribution capability and improving users' experience. To analyse the requests' patterns and fully utilize the universal cached contents, a novel intelligent resources management system is proposed, which enables effi cient cache resource allocation in real time, based on changing user demand patterns. The system is composed of two parts. The fi rst part is a fi ne-grain traffi c estimation algorithm called Temporal Poisson traffi c prediction(TP2) that aims at analysing the traffi c pattern(or aggregated user requests' demands) for different contents. The second part is a collaborative cache placement algorithm that is based on traffic estimated by TP2. The experimental results show that TP2 has better performance than other comparable traffi c prediction algorithms and the proposed intelligent system can increase the utilization of cache resources and improve the network capacity.展开更多
基金Natural Science Foundation of Shandong Province,China(ZR2019MD031)Philosophy and Social Science Planning Project of Tai’an City(2020skx035).
文摘Tai’an City is rich in tourism resources,and there are many A-class tourist attractions.Based on the functions and characteristics of geographic information system(GIS),as well as the spatial distribution characteristics and distribution structure of tourist attractions,the research methods of nearest-neighbor distance,connectivity,accessibility and closeness analysis are used to analyze road traffi c distribution between tourist attractions in Tai’an City.①The R value of the nearest-neighbor point is 0.27,which proves that tourist attractions in Tai’an City are in the state of clustering,and the spatial distance of tourist attractions in clustering area is small.②Theβindex andγindex of connectivity are 1.42 and 0.58,respectively,indicating that the traffi c network connectivity between tourist attractions is low,there is no dense traffi c network,and the optimal combination of tourism resources in tourist attractions is not high.③The accessibility index Ai is 37.68 km,indicating that the accessibility is low and the traffi c between tourist attractions is not perfect.The research results will provide the reference for scientifi c,systematic and objective comprehensive evaluation of existing A-class scenic areas and traffi c conditions in Tai’an City.
文摘The performance level of traffic control systems highly relies on accurate short-term network-wide traffic prediction.Signal processing techniques have been widely integrated with deep learning algorithms for this purpose,but no study has focused on how the parameter of wavelet decomposition order(level)affects the prediction robustness given the training data limitations.This study proposes a hybrid framework for short-term network-wide traffic state prediction,which applies multi-layer perceptron(MLP)neural networks to implicitly capture the traffic network co-movement patterns.Then,the framework is complemented by a seasonal auto-regressive integrated moving average(SARIMA)model,extracting the location-specific features,including the local seasonality and stochastic disturbances.Besides,the hybrid framework is used to explore the association between the traffic prediction accuracy and wavelet decomposition order using an order-adaptive discrete haar wavelet transform(DHWT).The proposed method was validated over four open-access datasets with different training data characteristics in Paris and Madrid urban areas.The results indicated that the hybrid framework significantly improved the predictive accuracy of the benchmark deep learning algorithms.The forecasts for the low-resolution dataset experienced a noticeable improvement once higher wavelet orders were applied.The statistical analysis revealed the moderating effects of the training sample size and data spatial resolution on the link between the wavelet decomposition orders and the model predictive performance.Nevertheless,a combination of pre-processing order of two,and post-processing order of three led to satisfactory results in the cases where sufficient historical data were available.
基金supported by the National High Technology Research and Development Program(863)of China(No.2015AA016101)the National Natural Science Fund(No.61300184)Beijing Nova Program(No.Z151100000315078)
文摘Information centric networking(ICN) is a new network architecture that is centred on accessing content. It aims to solve some of the problems associated with IP networks, increasing content distribution capability and improving users' experience. To analyse the requests' patterns and fully utilize the universal cached contents, a novel intelligent resources management system is proposed, which enables effi cient cache resource allocation in real time, based on changing user demand patterns. The system is composed of two parts. The fi rst part is a fi ne-grain traffi c estimation algorithm called Temporal Poisson traffi c prediction(TP2) that aims at analysing the traffi c pattern(or aggregated user requests' demands) for different contents. The second part is a collaborative cache placement algorithm that is based on traffic estimated by TP2. The experimental results show that TP2 has better performance than other comparable traffi c prediction algorithms and the proposed intelligent system can increase the utilization of cache resources and improve the network capacity.
