CNN-Based Spectral Super-Resolution of Panchromatic Night-Time Light Imagery: City-Size-Associated Neighborhood Effects

Data on artificial night-time light (NTL), emitted from the areas, and captured by satellites, are available at a global scale in panchromatic format. In the meantime, data on spectral properties of NTL give more information for further analysis. Such data, however, are available locally or on a commercial basis only. In our recent work, we examined several machine learning techniques, such as linear regression, kernel regression, random forest, and elastic map models, to convert the panchromatic NTL images into colored ones. We compared red, green, and blue light levels for eight geographical areas all over the world with panchromatic light intensities and characteristics of built-up extent from spatially corresponding pixels and their nearest neighbors. In the meantime, information from more distant neighboring pixels might improve the predictive power of models. In the present study, we explore this neighborhood effect using convolutional neural networks (CNN). The main outcome of our analysis is that the neighborhood effect goes in line with the geographical extent of metropolitan areas under analysis: For smaller areas, optimal input image size is smaller than for bigger ones. At that, for relatively large cities, the optimal input image size tends to differ for different colors, being on average higher for red and lower for blue lights. Compared to other machine learning techniques, CNN models emerged comparable in terms of Pearson’s correlation but showed performed better in terms of WMSE, especially for testing datasets.

The Neighborhood Effect on Keeping Non-Commuting Journeys within Compact and Sprawled Districts

The neighborhood effect on keeping non-commuting trips inside neighborhoods has not yet been investigated in developing countries. The modeling of non-commuting trips inside neighborhoods helps understand how to avoid unnecessary journeys by car into different parts of the city. This paper, therefore, attempts to clarify (1) the similarities and differences in the socioeconomic characteristics and the perceptions of people in sprawled and compact neighborhoods, (2) correlations between, on the one hand, the choice of destinations of non-commuting trips for shopping and entertainment activities and, on the other, the socioeconomic features, travel behavior, and perceptions of residents in the two large Pakistani cities of Lahore and Rawalpindi, (3) the similarities and differences in the determinants of non-commuting destinations inside neighborhoods in compact and sprawled districts. The paper develops four Binary Logistic (BL) regression models, with two models for each type of neighborhood. The findings show that trips to shopping areas inside compact districts are correlated with a sense of belonging to the neighborhood, frequency of public transport use, residential location, and mode choice of non-commuting trips to destinations both inside and outside the neighborhood. On the other hand, the number of non-commuting trips, mode choice for non-commuting trips outside the neighborhood, frequency of public transport use, the attractiveness of shops, and monthly income (please see the Note) are significant determinants for trips to the shopping area in sprawled districts. Age, gender, possession of a driver’s license, income, number of non-commuting trips, mode choice for non-commuting trips outside of the neighborhood, car ownership, and attractiveness of shops in a neighborhood are correlated with trips to entertainment locations inside the neighborhood in compact districts. Finally, the attractiveness of shops, quality of social and recreational facilities, a sense of belonging to a neighborhood, choice of residential location, gender, age, possession of a driver’s license, number of cars in the household, and income are determinants of trips to entertainment locations in sprawled districts. A chi-square test confirms the differences across gender, daily activity, monthly income, frequency of public transport use, residential location choice, and the quality of social and recreational facilities for sprawled and compact districts in Pakistan.

Life Stage of Neighbors Determines Neighborhood Effect on Tree Mortality in a Subtropical Forest

Neighborhood effects are a crucial ecological process that allow species to coexist in a forest. Conspecific and heterospecific neighbors, as major classified groups, affect tree mortality through various mechanisms associating with neighbor life stages. However, how neighbor life stages influence neighborhood effects and by what mechanisms remains a knowledge gap. Here we censused the mortality of 82,202 trees representing 30 species in a 20-ha subtropical forest and classified their neighbors into the following life stages: earlier, same and later. Then, we ran generalized linear mixed models to estimate the effect of neighbors at different life stages on tree mortality. Our results showed that conspecific later stage neighbors have effects on increasing tree mortality overall, whereas conspecific earlier stage neighbors have effects on decreasing. Furthermore, these opposing effects could offset each other so that the overall effect of conspecific neighbors on juvenile mortality seems small. In contrast, heterospecific neighbors have effects on decreasing tree mortality overall. These effects are consistent with those of later stage heterospecific neighbors. Our findings demonstrate that neighbors importantly impact tree mortality, and their specific effects are closely related to neighbor life stages. Any single effect from one neighbor life stage could disturb or dominate the total effects of the neighbors. Therefore, the neighbors must be divided into different life stages to best explain the neighborhood effect on forest dynamics.

An interpretable framework for investigating the neighborhood effect in POI recommendation

Geographical characteristics have been proven to be effective in improving the quality of point-of-interest (POI) recommendation. However, existing works on POI recommendation focus on cost (time or money) of travel for a user. An important geographical aspect that has not been studied adequately is the neighborhood effect, which captures a user’s POI visiting behavior based on the user’s preference not only to a POI, but also to the POI’s neighborhood. To provide an interpretable framework to fully study the neighborhood effect, first, we develop different sets of insightful features, representing different aspects of neighborhood effect. We employ a Yelp data set to evaluate how different aspects of the neighborhood effect affect a user’s POI visiting behavior. Second, we propose a deep learning–based recommendation framework that exploits the neighborhood effect. Experimental results show that our approach is more effective than two state-of-the-art matrix factorization–based POI recommendation techniques.

Permodelan Spasial pada Analisis Faktor yang Mempengaruhi Tingkat Pengangguran Terbuka Provinsi Bangka Belitung Tahun 2018

Unemployement is a multidimensional problem that have wide impact into progress and quality of one area. Based on that problem, it is necessary to have an analysis of factor that affected this phenomena. One economy phenomenon of one area can be influenced by neighborhood economy activity. The purpose of this study is to know factors that affected open unemployemnet rate also answer the problem of neighborhood effect by spatial model. Based on result, variables that having spatial effect are open unemployement rate, count of poor citizen, and also gross domestic product. Beside of that, it is also known that error spatial model is feasible to be a model because having smallest AIC score.

The Impact of Higher Education and Human Capital Quality on “Local-Neighborhood” Economic Growth

Based on China’s provincial panel data from 1990 to 2017 and the improved Lucas, Nelson & Phelps model, the Spatial Dubin Model is used to test the spatial effects of higher education and human capital quality. The results showed that high-level human capital, characterized by higher education and urban labor income index, indirectly promoted local economic growth through technological innovation. There was also a “local-neighborhood” synergy effect. The neighborhood effect was manifested in that it affected the economic development of neighbors by promoting technological catch-up. After considering the quality factor, both the local and neighborhood effects were enhanced. From a regional perspective, higher education in the Yangtze River Delta, where the level of economic development is relatively high, was manifested as a spatial spillover effect of technological innovation and the neighborhood effect in the northeastern Bohai Rim and the Pearl River Delta was manifested as a technological catch-up.