基於機器學習方法探討臺灣土地利用對於河川水質影響

、線上人數：145

、訪客IP：3.138.60.190

姓名	梁登傑(Deng-Jie Liang) 查詢紙本館藏	畢業系所	資訊工程學系
論文名稱	基於機器學習方法探討臺灣土地利用對於河川水質影響 (Exploring the Impact of Land Use on River Water Quality in Taiwan Based on Machine Learning Methods)
檔案	[Endnote RIS 格式] [Bibtex 格式] [相關文章] [文章引用] [完整記錄] [館藏目錄] 至系統瀏覽論文 ( 永不開放)
摘要(中)	過往研究中已發現土地利用與河川污染之間有著高度相關性，但目前並未有大規模的透過土地利用特徵預測河川污染物。本研究透過機器學習方法，基於內政部的土地利用資料和環境部的河川污染資訊，彌補此一空缺。資料涵蓋臺灣 54 個流域及 303 個測站，結合土地利用特徵與污染物數據進行模型訓練。研究透過因子分析篩選污染物，並應用聚類分析、隨機森林、極限梯度提升、支援向量機及人工神經網絡等模型，通過資料集切分評估準確性。結果顯示，對於特徵和污染物分群後再經由模型訓練，會比直接經由模型訓練結果更好。並且透過土地利用特徵和污染物採集年月預測污染物數值，可達到驗證集 R2（R squared）0.4 以上的結果，尤其是亞硝酸鹽氮的預測效果最佳。Shapley Value 評估指出土地利用特徵的重要性，多數人為土地利用與污染物濃度呈正相關，森林和水道沙洲灘地特徵呈負相關。本研究證明機器學習在環境污染物預測中的有效性，強調土地利用特徵的重要性，為未來污染控制和土地利用規劃提供科學依據。
摘要(英)	Previous studies identified a high correlation between land use and river pollution, but large-scale predictions of river pollutants based on land use characteristics are lacking. This research fills this gap using machine learning, with land use data from the Ministry of the Interior and river pollution data from the Ministry of Environment. The data covers 54 watersheds and 303 monitoring stations in Taiwan, combining land use characteristics and pollutant data for model training. The study employs factor analysis to screen pollutants and uses clustering analysis, random forest, extreme gradient boosting, support vector machine, and artificial neural networks to evaluate accuracy through cross-validation. Training the model after grouping features and pollutants yields better results than direct model training. Predicting pollutant values based on land use characteristics and the collection year and month can achieve an R-squared value of over 0.4 on the validation set, with nitrite nitrogen prediction performing the best. SHAP value assessment highlights the importance of land use characteristics, indicating a positive correlation between most anthropogenic land use and pollutant concentration, while forest and river sandbank features show a negative correlation.This study demonstrates the effectiveness of machine learning in predicting environmental pollutants, emphasizing the importance of land use characteristics, and providing a scientific basis for future pollution control and land use planning.
關鍵字(中)	★ 河川污染 ★ 土地利用 ★ 機器學習 ★ 夏普利值	關鍵字(英)	★ Water Pollution ★ Land use ★ Machine Learning ★ Shapley Value
論文目次	目錄摘要 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II 致謝 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III 目錄 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV 圖目錄 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI 表目錄 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XI 第一章緒論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 研究背景 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 文獻回顧 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 動機與目的 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 第二章材料和方法 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.1 研究地區 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.2 資料介紹 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2.1 污染物資料 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.2.2 土地利用資料 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2.3 人口密度資料 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.3 正規化和對數化 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.3.1 正規化 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.3.2 對數化 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.4 污染物資料篩選 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.5 資料集切分 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.6 機器學習模型 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.6.1 隨機森林 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.6.2 極限梯度提升 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.6.3 支援向量機 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.6.4 人工神經網路 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 2.6.5 超參數尋找 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.7 評估指標和 Shapley Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.7.1 評估指標 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.7.2 Shapley Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 第三章研究結果. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 3.1 污染物篩選結果 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 3.2 機器學習模型結果 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.3 Shapley Value 評估結果 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 第四章討論與結論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.1 討論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.2 結論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 參考文獻 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 附錄 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
參考文獻	參考文獻 [1] I. Zahoor and A. Mushtaq, “Water pollution from agricultural activities: A critical global review,” Int. J. Chem. Biochem. Sci, vol. 23, no. 1, pp. 164–176, 2023. [2] 環保署, 112 年版環境白皮書, https://www.moenv.gov.tw/DisplayFile.aspx? FileID=92DD7E871221D80B&P=276cd588-c66d-4031-93b2-3a011ec10fb3, [Accessed 2023]. [3] R. Srinivas, A. P. Singh, K. Dhadse, and C. Garg, “An evidence based integrated watershed modelling system to assess the impact of non-point source pollution in the riverine ecosys- tem,” Journal of Cleaner Production, vol. 246, p. 118 963, 2020. [4] S. Giri and Z. Qiu, “Understanding the relationship of land uses and water quality in twenty first century: A review,” Journal of Environmental Management, vol. 173, pp. 41–48, 2016. [5] R.-J. Chou, “Addressing watercourse sanitation in dense, water pollution-affected urban areas in taiwan,” Environment and Urbanization, vol. 25, no. 2, pp. 523–540, 2013. [6] X. Deng, “Influence of water body area on water quality in the southern jiangsu plain, eastern china,” Journal of Cleaner Production, vol. 254, p. 120 136, 2020. [7] L.-C. Chiang, Y.-C. Wang, Y.-K. Chen, and C.-J. Liao, “Quantification of land use/land cover impacts on stream water quality across taiwan,” Journal of Cleaner Production, vol. 318, p. 128 443, 2021. [8] M. Zhu et al., “A review of the application of machine learning in water quality evaluation,” Eco-Environment & Health, vol. 1, no. 2, pp. 107–116, 2022. [9] R. Wang, J.-H. Kim, and M.-H. Li, “Predicting stream water quality under different urban development pattern scenarios with an interpretable machine learning approach,” Science of the Total Environment, vol. 761, p. 144 057, 2021. [10] M. G. Zanoni, B. Majone, and A. Bellin, “A catchment-scale model of river water quality by machine learning,” Science of the Total Environment, vol. 838, p. 156 377, 2022. [11] H. I. H. Yusri, S. L. M. Hassan, I. S. A. Halim, N. E. Abdullah, et al., “Water quality classi- fication using svm and xgboost method,” in 2022 IEEE 13th Control and System Graduate Research Colloquium (ICSGRC), IEEE, 2022, pp. 231–236. [12] A. H. Haghiabi, A. H. Nasrolahi, and A. Parsaie, “Water quality prediction using machine learning methods,” Water Quality Research Journal, vol. 53, no. 1, pp. 3–13, 2018. [13] 行政院, 國情介紹-土地與人民, https://www.ey.gov.tw/state/4447F4A951A1EC45/ 094b1d53-de8d-4393-bde6-ab092969cce4, [Accessed 29-03-2024]. [14] 環境部全國環境水質監測資訊網, 歷史數據, https://wq.moenv.gov.tw/EWQP/zh/ ConService/DownLoad/HistoryData.aspx, [Accessed 31-05-2024]. [15] 全國法規資料庫, 地面水體分類及水質標準, https://law.moj.gov.tw/LawClass/ LawAll.aspx?pcode=O0040005, [Accessed 31-05-2024]. [16] 全國法規資料庫, 自來水水質標準, https://law.moj.gov.tw/LawClass/LawSingle. aspx?pcode=J0110037&flno=5, [Accessed 11-06-2024]. [17] 環境部, 全國環境水質監測資訊網, https : / / wq . moenv . gov . tw / EWQP / zh / Encyclopedia/NounDefinition/Pedia_37.aspx, [Accessed 11-06-2024]. [18] 內政部國土測繪中心, 第二次國土利用調查分類系統表 (95 年度), https://www. nlsc.gov.tw/cp.aspx?n=13721, [Accessed 06-06-2024]. [19] A. G. Yong, S. Pearce, et al., “A beginner’s guide to factor analysis: Focusing on exploratory factor analysis,” Tutorials in Quantitative Methods for Psychology, vol. 9, no. 2, pp. 79–94, 2013. [20] scikit-learn, Randomforestregressor, https://scikit- learn.org/stable/modules/ generated / sklearn . ensemble . RandomForestRegressor . html, [Accessed 07-06- 2024]. [21] E. Alpaydin, Introduction to Machine Learning. MIT press, 2020. [22] T. Chen and C. Guestrin, “Xgboost: A scalable tree boosting system,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794. [23] Xgboost python package, https://xgboost.readthedocs.io/en/stable/python/, [Accessed 09-07-2024]. [24] scikit-learn, Svr, https://scikit-learn.org/stable/modules/generated/sklearn. svm.SVR, [Accessed 07-06-2024]. [25] S. Haykin, Neural Networks and Learning Machines, 3/E. Pearson Education India, 2009. [26] Tensorflow — pypi.org, https://pypi.org/project/tensorflow/2.16.1/, [Accessed 07-06-2024]. [27] B. Shahriari, K. Swersky, Z. Wang, R. P. Adams, and N. De Freitas, “Taking the human out of the loop: A review of bayesian optimization,” Proceedings of the IEEE, vol. 104, no. 1, pp. 148–175, 2015. [28] A. E. Roth, The Shapley value: essays in honor of Lloyd S. Shapley. Cambridge University Press, 1988. [29] SHAP, Api reference, https : / / shap - lrjball . readthedocs . io / en / latest / api . html, [Accessed 08-06-2024]. [30] B. Boehmke and B. M. Greenwell, Hands-On Machine Learning with R. Chapman and Hall/ CRC, 2019. 73
指導教授	洪炯宗陳瑞昇(Jorng-Tzong Horng Jui-Sheng Chen)	審核日期	2024-7-29
推文	facebook plurk twitter funp google live udn HD myshare reddit netvibes friend youpush delicious baidu
網路書籤	Google bookmarks del.icio.us hemidemi myshare

博碩士論文 111522037 詳細資訊