The ability to rapidly produce accurate land use and land cover maps regularly and consistently has been a growing initiative as they have increasingly become an important tool in the efforts to evaluate, monitor, and conserve Earth’s natural resources. Algorithms for supervised classification of satellite images constitute a necessary tool for the building of these maps and they have made it possible to establish remote sensing as the most reliable means of map generation. In this paper, we compare three machine learning techniques: Random Forest, Support Vector Machines, and Light Gradient Boosted Machine, using a 70/30 training/testing evaluation model. Our research evaluates the accuracy of Light Gradient Boosted Machine models against the more classic and trusted Random Forest and Support Vector Machines when it comes to classifying land use and land cover over large geographic areas. We found that the Light Gradient Booted model is marginally more accurate with a 0.01 and 0.059 increase in the overall accuracy compared to Support Vector and Random Forests, respectively, but also performed around 25% quicker on average.
This study develops multiple evaluation indexes in the context of sustainable urban regeneration through introducing smart technologies/infrastructures and assesses 63 local urban regeneration strategic plans by using the content analysis method. A total of 107 indexes are developed based on the four aspects (economy, society and culture, environment, and livability) of sustainability. From our findings, the average plan quality score of 54 local governments’ plans is 17.5 out of 50, with the metropolitan governments’ plans averaging 16.8, which indicates that the plans currently sampled do not sufficiently reflect the basic concepts of sustainable and smart urban regeneration. The contents of most of the plans generally focus on specific sectors, such as society, culture, and housing, whereas smart technology-related information and policies are relatively deficient. Among the five plan components (factual bases, goals/objectives, policies/strategies, implementation, coordination) reviewed, the implementation component receives the highest score, while indicators related to action strategies are mentioned least often. In particular, the results reveal that indexes relating to the energy and transportation sectors are not frequently mentioned; as such, each municipality is recommended to work to increase awareness of smart technologies and policies. For urban regeneration projects to be sustainable, multi-faceted policies must be implemented by various stakeholders with a long-term perspective. The results of this study can be used as a base for local planners and decision-makers when adopting and supplementing existing regeneration plans, and can contribute to promoting more sustainable urban regeneration through actively adopting various smart technologies initiatives.
Given the rapidly increasing need for policies with regard to single-person households in Korea, this study examines the effects of park accessibility and the connectivity of green spaces on the spatial distribution of single-person households. SK-Tmap API and Conefor 2.6 are used to analyze park accessibility and green space connectivity, respectively. Multiple and spatial regression analyses are conducted using variables for the following three characteristics: park and green space, housing, and region. The findings show that generalized Betweenness Centrality–Integral Index of Connectivity based index (dBC_IIC), apartments, studio apartments, housings larger than 85 m2, distance to welfare facilities, and population density had a positive association with the spatial distribution of single-person households, while park accessibility, difference in Number of Links (dNL), generalized Betweenness Centrality–Probability of Connectivity based index (dBC_PC), and housing sale prices had a negative relationship. Regression analyses are further conducted for different age groups (10–20 years, 30–50 years, and over 60 years). In terms of park connectivity, dBC_PC showed a negative effect and dBC_IIC had a positive effect for the 10–20 age groups, while the 30–50 age group showed the same result as that of all single-person households. For single-person households over 60 years of age, no connectivity index was found to be significant. Policy implications are made in the short- and mid- to long-term for strengthening the connectivity of parks and green spaces in the study area. The results of this study can be used as an important guideline for establishing park and green space plans in consideration of single-person households in the future.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.