共享單車的時空資料記錄了豐富的人文活動資訊，得以反映不同的土地使用類別在時空上相應的交通流動模式，由此確立了土地使用和共享單車使用模式間相互印證的可行性。土地使用分類一直是都市規劃和土地利用管理中的重要問題，能夠準確地進行都市土地分類可以幫助決策者更好地制定都市發展計劃和政策。近年來隨著大數據普遍容易取得和人工智慧的快速發展，許多研究利用傳統機器學習或是深度學習結合衛星影像，進行大面積土地使用分類。雖然衛星影像在區分光譜特性差異明顯的建物、水體、植被下有相當好的表現，但若以使用分區對建物做細分，只靠衛星影像可能無法提供足夠的辨識資訊，而社會感知資料(Social Sensing Data)可反映人類的活動模式，藉以補足衛星影像的不足。因此，本研究以共享單車騎乘數據為社會感知資料，結合遙測影像，利用傳統機器學習和深度學習模型分別進行以點、區塊、像素為單位之都市土地使用分類預測。預測結果以資料內容而言，結合遙測與社會感知資料整體精度最佳，只使用社會感知資料次之，只使用遙測影像最差。以資料單位而言，像素分類整體精度優於區塊分類和點分類。綜上所述，各組數據排列組合下以使用 XGBoost 模型，以遙測資料結合社會感知資料進行像素分類的預測表現最佳，整體精度為 0.89，論整體精度和模型穩健性(Robustness)以隨機森林表現最佳，本研究也發展都市發展時空模擬模型，模擬未來 5 年後和 10 年後商業區和住宅區的發展位置。透過本研究印證了共享單車站點租借和歸還人次於時空分布之特性和都市土地分類的空間分布有密切關係，都市土地使用分類可說是依據人文活動來劃分使用分區。利用共享單車騎乘資料的時間與空間特性進行都市土地使用分類預測，可以為都市規劃和土地管理提供客觀真實之案例應用。

Urban land use classification is a crucial aspect of urban planning and land management.Accurate classification can assist policymakers in developing better urban developmentplans and policies. With the advent of readily available satellite imagery and rapid advancements in artificial intelligence, many studies have leveraged machine learning or deep learning combined with satellite imagery for large-scale land use classification. While satellite images excel at distinguishing buildings, water bodies, and vegetation due to their spectral characteristics, they often fall short in providing sufficient information for detailed classification based on building usage. Social sensing data, reflecting human activity patterns, can complement satellite imagery in this regard. This study utilizes bike-sharing data as social sensing data, combined with remote sensing images, to predict urban land use classification at point, area, and pixel levels using machine learning and deep learning models. The study reveals that integrating remote sensing data with social perception data provides the highest overall accuracy, followed by using only social perception data, with remote sensing data alone yielding the lowest accuracy. When considering data units, pixel-based classification performs better than area-based and point-based classifications. Among the models used, machine learning models, specifically XGBoost, outperform deep learning models in terms of overall accuracy.This study also develops a spatiotemporal simulation model for urban development, simulating the locations of commercial and residential areas 5 and 10 years into the future.

李舒媛. (2018). 以悠遊卡大數據探討 YouBike 租賃及轉乘捷運之使用者行為. 淡江大學運輸管理學系運輸科學碩士班學位論文, 2018, 1-105.
孔繁恩, 詹進發, 邵怡誠, 李茂園, 葉堃生, & 陳連晃. (2014). 物件式分類法於高解析度航照影像萃取崩塌地之研究. 航測及遙測學刊, 18(4), 267-281.
謝雅婷. (2011). 以 SEED CA 模型探討台南市五期重劃區之建地成長特性 (Doctoral dissertation, National Cheng Kung University Department of Urban Planning).
Youbike 微 笑 單 車 。 臺 南 市 地 區 租 賃 次 數 總 數(2024 年 5 月)。https://www.youbike.com.tw/region/tainan/operation/
Anugraha, A. S., & Chu, H. J. (2018). Land use classification from combined use of remote sensing and social sensing data. The International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences, 42, 33-39.
Abdi, A. M. (2020). Land cover and land use classification performance of machine learning algorithms in a boreal landscape using Sentinel-2 data. GIScience & Remote Sensing, 57(1), 1-20.
Alshari, E. A., Abdulkareem, M. B., & Gawali, B. W. (2023).Classification of landuse/land cover using artificial intelligence (ANN-RF). Frontiers in Artificial Intelligence,5, 964279.
Baroud, S., Chokri, S., Belhaous, S., Hidila, Z., & Mestari, M. (2020, June). An artificial neural network combined to object oriented method for land cover classification of high resolution RGB remote sensing images. In International Conference on Smart Applications and Data Analysis (pp. 221-232). Cham: Springer International Publishing.
Boston, T., Van Dijk, A., Larraondo, P. R., & Thackway, R. (2022). Comparing CNNs and random forests for Landsat image segmentation trained on a large proxy land cover dataset.Remote Sensing, 14(14), 3396.
Campbell, A. A., Cherry, C. R., Ryerson, M. S., & Yang, X. (2016). Factors influencing the choice of shared bicycles and shared electric bikes in Beijing. Transportation research part C: emerging technologies, 67, 399-414.
Cheng, H. H., & Hsu, Y. Y. (2022). Integrating spatial multi-criteria evaluation into the potential analysis of culture-led urban development–A case study of Tainan. Environment and Planning B: Urban Analytics and City Science, 49(1), 335-351.
Citi Bike.Citi Bike system data(n.d.). https://citibikenyc.com/system-data
Dong, X., Xu, Y., Huang, L., Liu, Z., Xu, Y., Zhang, K., ... & Wu, G. (2020). Exploring impact of spatial unit on urban land use mapping with multisource data. Remote Sensing,12(21), 3597.
European Space Agency.S2 Processing (n.d.).https://sentiwiki.copernicus.eu/web/s2-processing
Ghayour, L., Neshat, A., Paryani, S., Shahabi, H., Shirzadi, A., Chen, W., ... & Ahmad, A.(2021). Performance evaluation of sentinel-2 and landsat 8 OLI data for land cover/use classification using a comparison between machine learning algorithms. Remote Sensing, 13(7), 1349.
Ji, Y., Fan, Y., Ermagun, A., Cao, X., Wang, W., & Das, K. (2017). Public bicycle as a feeder mode to rail transit in China: The role of gender, age, income, trip purpose, and bicycle theft experience. International Journal of Sustainable Transportation, 11(4), 308-317.
Kadavi, P. R., & Lee, C. W. (2018). Land cover classification analysis of volcanic island in Aleutian Arc using an artificial neural network (ANN) and a support vector machine(SVM) from Landsat imagery. Geosciences Journal, 22, 653-665.
Mishra,V.N., Prasad, R., Kumar, P., Gupta, D. K., & Srivastava, P. K. (2017). Dual-polarimetric C-band SAR data for land use/land cover classification by incorporatingtextural information. Environmental Earth Sciences, 76, 1-16.
Ma, X., Ji, Y., Yang, M., Jin, Y., & Tan, X. (2018). Understanding bikeshare mode as a feeder to metro by isolating metro-bikeshare transfers from smart card data. Transport policy, 71, 57-69.
Shi, Y., Qi, Z., Liu, X., Niu, N., & Zhang, H. (2019). Urban land use and land cover classification using multisource remote sensing images and social media data. Remote Sensing, 11(22), 2719.
Sun, C., & Lu, J. (2023). The relative roles of different land-use types in bike-sharing demand: A machine learning-based multiple interpolation fusion method. Information Fusion, 95, 384-400.
Tang, Y., Pan, H., & Shen, Q. (2011, January). Bike-sharing systems in Beijing, Shanghai,and Hangzhou and their impact on travel behavior. In Transportation Research Board 90th Annual Meeting (Vol. 11, p. 3862).
Hu, T., Yang, J., Li, X., & Gong, P. (2016). Mapping urban land use by using landsat images and open social data. Remote sensing, 8(2), 151.
Wang, X., Gao, X., Zhang, Y., Fei, X., Chen, Z., Wang, J., ... & Zhao, H. (2019). Landcover classification of coastal wetlands using the RF algorithm for Worldview-2 and Landsat 8 images. Remote Sensing, 11(16), 1927.
Wang, X., Cheng, Z., Trépanier, M., & Sun, L. (2021). Modeling bike-sharing demand using a regression model with spatially varying coefficients. Journal of Transpor tGeography, 93, 103059.
Wang, S., Fei, T., Li, W., Zhang, A., Guo, H., & Du, Y. (2022). Incorporation of intra-city human mobility into urban growth simulation: A case study in Beijing. Journal of Geographical Sciences, 32(5), 892-912.
Xu, D. (2020). Free wheel, free will! the effects of bikeshare systems on urban commuting patterns in the us. Journal of Policy Analysis and Management, 39(3), 664-685.
Zhang, Y., Thomas, T., Brussel, M., & Van Maarseveen, M. (2017). Exploring the impact of built environment factors on the use of public bikes at bike stations: Case study in Zhongshan, China. Journal of transport geography, 58, 59-70.