[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126764-en":3,"doc-seo-126764-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},126764,962084928432,"Emma Wilson","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Exploring Machine Learning Approaches to Precipitation Prediction - Post Processing of Daily Accumulated North American Forecasts","This thesis investigates recent machine learning and deep learning approaches to improve 24-hour precipitation forecasts through post-processing. It leverages a comprehensive North American dataset combining precipitation values derived from Numerical Weather Prediction (NWP) models with additional meteorological features. The study demonstrates clear gains over a baseline input-means model, reporting substantial error reductions, including 15% lower MAE, 5% lower RMSE, 45% lower MdAE, and a 50% decrease in relative bias. Convolutional Neural Networks and Gradient Boosting Regressor deliver the strongest predictive performance for daily precipitation.","Exploring Machine Learning Approaches to Precipitation Prediction: Post Processing of Daily Accumulated North American forecasts  \nby  \nRushil Goomer  \nA Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of  \nMaster of Science  \nin the Department of Applied Computer Science  \nRushil Goomer, 2023  \nUniversity of Winnipeg  \nii  \nExploring Machine Learning Approaches to Precipitation Prediction: Post Processing of Daily Accumulated North American forecasts  \nby  \nRushil Goomer  \nSupervisory Committee  \n\n| Dr. S. Ramanna, Supervisor\u003Cbr>(Department of Applied Computer Science) |\n| --- |\n| Dr. C. Valdemerra, Member\u003Cbr>(Department of Applied Computer Science) |\n\nDr. K. Kotecha, External Member  \n(Symbiosis Centre for Applied Artificial Intelligence (SCAAI), Symbiosis International, Pune, India)  \niii  \nABSTRACT  \nThis thesis presents recent work on exploring machine learning (ML) and deep learning (DL) models to improve the accuracy of 24 hour precipitation forecasts. Leveraging a comprehensive North American dataset of precipitation values from Numerical Weather Prediction (NWP) models and secondary meteorological features, the research showcases the need of ML techniques in post-processing NWP precipitation predictions. The evaluation reveals remarkable performance improvements over baseline model, with certain ML models achieving a 15% reduction in Mean Absolute Error (MAE), a 5% decrease in Root Mean Squared Error (RMSE), a 45% reduction in Median Absolute Error (MdAE), and a 50% decrease in Relative Bias (RB) . Convolutional Neural Networks (CNN) and Gradient Boosting Regressor (GBR) emerged as top performers, demonstrating their proficiency in accurately predicting daily precipitation.  \nKeywords: Numerical Weather Prediction (NWP), Precipitation Forecasting, Machine Learning, Neural Networks, Gradient Boosting, Graph Neural Networks, Weather Forecast, Post-processing, Meteorological Features.  \niv  \nContents  \nSupervisory Committee ii  \nAbstract iii  \nTable of Contents iv  \nList of Tables vii  \nList of Figures viii  \nAcknowledgements xi  \nDedication xii  \n1 Introduction 1  \n1.1 Problem Statement and Motivation ................... 2  \n1.2 Previous Work .............................. 3  \n1.3 Proposed Research ............................ 5  \n1.4 Thesis Outline ............................... 6  \n2 Related Works 7  \n2.1 Weather Prediction and Machine Learning ............... 7  \n2.2 Post Processing NWP models with Machine Learning ......... 9  \n3 Preliminaries 12  \n3.1 Introduction ................................ 12  \n3.2 Input Means Model (Baseline) ...................... 12  \n3.3 Machine Learning Fundamentals ..................... 13  \n3.4 Multiple Linear Regression ........................ 13  \n3.5 Ensemble Methods ............................ 14  \nv  \n3.5.1 Random Forest .......................... 15  \n3.5.2 Gradient Boosting Regression (GBR) .............. 16  \n3.5.3 Extreme Gradient Boost (XGboost) ............... 17  \n3.6 Neural Networks ............................. 19  \n3.6.1 Neural Networks (NN) ...................... 20  \n3.6.2 Convolutional Neural Networks (CNN) ............. 22  \n3.6.3 Graph Convolutional Neural Networks (GNN) ......... 26  \n3.6.4 Evaluation Metrics ........................ 28  \n4 Data Procuring, Curating, and Preparing 31  \n4.1 Geographical Area Covered ....................... 31  \n4.2 Input Weather Model Details ...................... 32  \n4.3 Dataset Acquisition ............................ 36  \n4.4 Data Preparation and Augmentation .................. 36  \n4.5 Feature Selection ............................. 40  \n5 Results and Discussion 45  \n5.1 Mean Absolute Error (MAE) ....................... 46  \n5.2 Root Mean Squared Error (RMSE) ................... 48  \n5.3 Median Absolute Error (MdAE) ..................... 48  \n5.4 Maximum Error (MaxE) ......................... 51  \n5.5 Correlation Coefficient (CC) ....................... 52  \n5.6 Relative Bias (RB) ..........","cbCaitzmfJt2fpBn","https://ap.wps.com/l/cbCaitzmfJt2fpBn","pdf",18734213,1,95,"English","en",105,"# 1 Introduction\n## 1.1 Problem Statement and Motivation\n## 1.2 Previous Work\n## 1.3 Proposed Research\n## 1.4 Thesis Outline\n# 2 Related Works\n## 2.1 Weather Prediction and Machine Learning\n## 2.2 Post Processing NWP Models with Machine Learning\n# 3 Preliminaries\n## 3.2 Input Means Model (Baseline)\n## 3.3 Machine Learning Fundamentals\n## 3.4 Multiple Linear Regression\n## 3.5 Ensemble Methods\n## 3.6 Neural Networks\n# 4 Data Procuring, Curating, and Preparing\n## 4.1 Geographical Area Covered\n## 4.2 Input Weather Model Details\n## 4.3 Dataset Acquisition\n## 4.4 Data Preparation and Augmentation\n## 4.5 Feature Selection\n# 5 Results and Discussion\n## 5.1 Mean Absolute Error (MAE)\n## 5.2 Root Mean Squared Error (RMSE)\n## 5.3 Median Absolute Error (MdAE)\n## 5.4 Maximum Error (MaxE)\n## 5.5 Correlation Coefficient (CC)\n## 5.6 Relative Bias (RB)\n## 5.7 Probability of Detection (POD)\n## 5.8 False Alarm Ratio (FAR)\n## 5.9 Critical Success Index (CSI)\n## 5.10 Confusion Matrix (CM)\n## 5.11 Discussion\n# 6 Conclusions\n# Bibliography\n# A Appendix","[{\"question\":\"What problem does the thesis address in precipitation prediction?\",\"answer\":\"The thesis targets improving the accuracy of 24-hour precipitation forecasts by applying machine learning and deep learning models as post-processing to NWP outputs.\"},{\"question\":\"What data sources are used for training and evaluation?\",\"answer\":\"It uses a North American dataset that combines precipitation values from NWP numerical weather models with additional meteorological features to construct model inputs.\"},{\"question\":\"Which models performed best and what improvements were reported?\",\"answer\":\"Convolutional Neural Networks (CNN) and Gradient Boosting Regressor (GBR) showed top performance, achieving notable reductions in multiple error metrics, including 15% lower MAE and 5% lower RMSE versus the baseline.\"}]","Exploring Machine Learning Approaches to Precipitation Prediction - Post Processing of Daily Accumulated North American Forecasts | PDF",1785934666,239,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"exploring-machine-learning-approaches-to-precipitation-prediction-post-processing-of-daily-accumulated-north-american-forecasts","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/exploring-machine-learning-approaches-to-precipitation-prediction-post-processing-of-daily-accumulated-north-american-forecasts/126764/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the thesis address in precipitation prediction?","Question",{"text":75,"@type":76},"The thesis targets improving the accuracy of 24-hour precipitation forecasts by applying machine learning and deep learning models as post-processing to NWP outputs.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What data sources are used for training and evaluation?",{"text":80,"@type":76},"It uses a North American dataset that combines precipitation values from NWP numerical weather models with additional meteorological features to construct model inputs.",{"name":82,"@type":73,"acceptedAnswer":83},"Which models performed best and what improvements were reported?",{"text":84,"@type":76},"Convolutional Neural Networks (CNN) and Gradient Boosting Regressor (GBR) showed top performance, achieving notable reductions in multiple error metrics, including 15% lower MAE and 5% lower RMSE versus the baseline.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]