Fig 1.
Research process.
Fig 2.
(a) Central and Southeastern districts in Seoul. (1) Jongro-gu; (2) Jung-gu; (3) Yongsan-gu; (4) Seocho-gu; (5) Gangnam-gu; (6) Songpa-gu; (7) Gangdong-gu. The central area includes areas 1-3, and the southeastern area includes areas 4-7; (b) Spatial characteristics of Seoul, including land uses and zoning areas; Source: Ministry of the Interior and Safety (http://www.juso.go.kr/externalLink/goUrl.do?menuId=DT05) and Seoul Metropolitan Government (https://data.seoul.go.kr/dataList/OA-21127/S/1/datasetView.do, https://data.seoul.go.kr/dataList/OA-21133/S/1/datasetView.do, https://data.seoul.go.kr/dataList/OA-21136/S/1/datasetView.do, https://data.seoul.go.kr/dataList/OA-21137/S/1/datasetView.do). These public works are used according to Korea Open Government License (KOGL).
Table 1.
Acquired texts by district.
Fig 3.
Urban soundscape taxonomy adapted from the place model.
This consists of a sense of place, including physical setting, activity, and meaning (image).
Fig 4.
The result of Elbow method and silhouette measurement.
(a) Elbow method shows that the rate of decrease in WCSS (inertia) slows significantly after =3; (b) Silhouette score shows the highest score at
=3.
Table 2.
Descriptive statistics of the target areas.
Fig 5.
POIs type distribution by neighborhood group.
This illustrates the difference in POI types between the central area and the southeastern area. Particularly, the central area, which has a longer history compared to the southeastern area, was found to have more attraction-related POIs (i.e., historical, tourism, culture, shopping, and entertainment) (Mann-Whitney test, p < 0.001).
Table 3.
Soundscape keyword ranking (Top 20) in collected texts from social media.
Fig 6.
Dominant urban soundscapes by district.
This map depicts the target area in 1.5 × 1.5 grid cell units to identify the spatial distribution of soundscape types. It shows the primary soundscape categories for 77 sub-regions, specifically the category with the highest proportion. Where categories have the same proportion, they are represented as two or more; Source: Ministry of the Interior and Safety (http://www.juso.go.kr/externalLink/goUrl.do?menuId=DT05) and Seoul Metropolitan Government (https://data.seoul.go.kr/dataList/OA-21136/S/1/datasetView.do). These public works are used according to Korea Open Government License (KOGL).
Fig 7.
Proportion of urban soundscapes by district.
It displays the proportion and composition of soundscape categories in each district. Activity-related soundscapes are dominant across all districts.
Fig 8.
Pearson correlation matrix of soundscape categories.
This shows correlations between soundscape categories. Especially, indoor-mechanical (r = −0.46), mechanical-behavior (r = −0.45), and indoor-behavior (r = −0.39) soundscapes exhibit relatively strong negative correlations. The Memory category was excluded from the analysis because no keywords were found in all regions.
Fig 9.
Urban soundscape patterns by district.
This map shows 77 sub-regions classified into three distinct patterns (Type 1, Type 2, and Type 3) based on K-means clustering results. To compare the spatial distribution of soundscape patterns, the target area is represented as 1.5 × 1.5 grid cell units; Source: Ministry of the Interior and Safety (http://www.juso.go.kr/externalLink/goUrl.do?menuId=DT05) and Seoul Metropolitan Government (https://data.seoul.go.kr/dataList/OA-21136/S/1/datasetView.do). These public works are used according to Korea Open Government License (KOGL).
Fig 10.
Distributions of the proportion of soundscape in each cluster.
This box plot shows the mean proportion of soundscape categories for each cluster corresponding to the areas. (a) Type 1; (b) Type 2; (c) Type 3. X-axis: Soundscape types; Y-axis: Soundscape proportion.
Table 4.
The results of ANOVA and Tukey’s HSD.