Search

To tackle the issue of traffic data loss due to various software/hardware failures in urban road network environments, this paper proposed a traffic data imputation method based on spatial-temporal residual tensor learning (ST-RTL）. This method constructed a 3D traffic tensor with missing value to characterize original spatiotemporal attributes of road network maxi-mally. Then it adopted Gibbs sampling to perform a CANDECOMP/PARAFAC (CP) tensor decomposition and low-rank reconstruction of missing traffic data based on the assumption of Gaussian distribution. Considering the residual value produced by the tensor repair process, the study designed a bidirectional residual optimization structure with dynamic iterations to capture the residual spatiotemporal dependencies to enable the accurate repair of the missing traffic data. The experiments took a publicly available Hangzhou metro passenger flow for model construction and validation. The results indicate that when the missing rates are 10%～80%, the three missing scenarios (random, cluster and hybrid missing) have large differences on tensor structure damage, among which cluster missing has the greatest destruction and the evaluation indexes MAPE, RMSE and MAE of ST-RTL lied in 3.1071～7.0371, 16.3779～58.4286 and 3.7434～8.0135; and each indicator of ST-RTL model shows an accelerated increasing trend as the missing rate rises. Compared with the representative baseline models such as HaLRTC, GAIN and BGCP, the ST-RTL exhibits lower performance metrics and stronger stability in the acceptable computational costs, which can provide high-quality basic data for intelligent transportation systems. [ABSTRACT FROM AUTHOR]

The technological innovation of emerging industries is the key field of regional competition, the impact of the rapid development of regional innovation integration on localization is a question that economic geographers need to answer. Taking the new material industry in the Yangtze River Delta as an example, this paper obtains the data of invention patents from Incopat patent service website, and explores the relationship between regional innovation integration and localization in cooperative R & D and patent transfer network from 2000 to 2017. The main conclusions are as follows: (1) In terms of spatial structure, the degree of regional integration in the two networks is highly related to the spatial pattern of localization, and both of them present a "core periphery" structure centered on the "Z" shaped region composed of "Hefei - Nanjing - Shanghai - Hangzhou - Ningbo". (2) The degree of regional innovation integration can promote the strength of local network, but there is a significant positive "U" relationship between integration and localization. On the one hand, it affirms that the regional innovation integration has positive significance for the improvement of regional overall innovation ability. On the other hand, it shows that the role of de-localization is dominant in the early regional innovation integration. But after a certain degree, the process of high-quality integration can force the local network to accelerate the development, and the regions with high degree of integration can take the lead in exerting the siphon effect and utilizing external resources to promote the process of re-localization. (3) The process of regional innovation integration is an interwoven process of de-localization and re-localization. It is crucial for local governments to reshape local networks and local characteristics with the help of integration forces. [ABSTRACT FROM AUTHOR]

Objectives: As the acceleration of global urbanization, traffic congestion and environmental degradation have emerged in the urban area. Based on the light and statistical data, this paper studied the spatial evolution between urban people and land, and revealed the significance of its coupling and coordination characteristics for the sustainable development of cities. Methods: For primary data, a series of preprocessing are needed to eliminate the problems of supersaturation and discontinuity of inter‑annual data. Based on the lighting threshold, this paper excerpted the built‑up areas of urban agglomeration in Zhejiang Province year by year, compared them with the statistical data, and analyzed the corrected data source relatively. Firstly, based on the order‑scale distribution model of urban order and urban population in Zhejiang Province from 1992 to 2020, we made a dynamic analysis on population aggregation and distribution in its development process. Then, we constructed the population‑land urbanization allometric growth model by using population and light area, as well as judged the spatialization of man‑land relationship in urban development. Finally, according to the lag phenomenon of population growth relative to land development, we proposed an allometric growth model with lag factor, and used the lag term model to judge the characteristics of unequal growth of people and land in different cities. Results: The structural system of urban agglomeration is relatively mature in Zhejiang Province, and the distribution of cities is reasonable. However, the primacy of city is gradually improved, and the population is gradually gathering. The results of allometric growth model show that the man‑land relationship of urban agglomeration in Zhejiang Province is generally low in the eastern coastal cities and high in the western inland cities, and the planning of the four metropolitan areas is reasonable. The improvement of fitting accuracy of the proposed model with lag factor obviously shows that there is lag effect between population urbanization and land urbanization, with an average increase of 6.47% and the highest increase of 12.61%. Among the influencing factors on land urbanization, the dominant factor in Hangzhou is the existing population in urban areas, while the dominant factor in Ningbo is the new population in urban areas. Conclusions: The proposed allometric growth model with lag factor can correct the model errors and improve the fitting accuracy of urban man‑land relationship. It can also effectively evaluate the temporal and spatial evolution characteristics of man‑land relationship in regional system, explore its development and put forward countermeasures. [ABSTRACT FROM AUTHOR]