[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122049-en":3,"doc-seo-122049-105":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":20,"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},122049,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","A Novel Spatial Bagging Algorithm for Predictive Ensemble Machine Learning - Master of Science Thesis","A novel spatial bagging algorithm is developed to improve statistical inference for geoscientific decision making in settings with spatially correlated data. Existing machine learning methods assume independent samples, so the approach models spatial dependence using variograms and unconditional simulations, then derives an effective sample size to construct bootstrapped samples that behave as independent data. Prediction accuracy and uncertainty quantification are validated on extensive synthetic datasets with varied correlation and noise, and on a real-world case study. Results show robustness to overfitting and substantially better uncertainty modeling than standard bagging, with equal prediction accuracy in noise-free cases and superior performance as noise increases.","Copyright by  \nFehmi Ozbayrak 2024  \n1  \nThe Thesis Committee for Fehmi Ozbayrak certifies that this is the approved version of the following thesis:  \nA Novel Spatial Bagging Algorithm for Predictive Ensemble  \nMachine Learning  \nSUPERVISING COMMITTEE:  \nMichael James Pyrcz, Supervisor  \nJohn Timothy Foster  \nA Novel Spatial Bagging Algorithm for Predictive Ensemble  \nMachine Learning  \nby  \nFehmi Ozbayrak  \nThesis  \nPresented to the Faculty of the Graduate School of The University of Texas at Austin  \nin Partial Fulfillment  \nof the Requirements  \nfor the Degree of  \nMaster of Science in Engineering  \nThe University of Texas at Austin May 2024  \nDedication  \nTo my mother.  \nAbstract  \nA Novel Spatial Bagging Algorithm for Predictive Ensemble  \nMachine Learning  \nFehmi Ozbayrak, MSE  \nThe University of Texas at Austin, 2024  \nSUPERVISOR: Michael James Pyrcz  \nStatistical inference has become critical for decision making in geoscientific industries, including exploration, oil and gas production, mineralogy, and environmental assessment, with the emergence of machine learning and data-driven approaches. The methods of these areas are developed and matured for independent data, however, spatial data is of great interest to geoscience, and it is a beneficial endeavor to adapt the existing machine learning methods to spatially correlated data. We hypothesize a novel bagging algorithm for spatially dependent data, where we capture the spatial correlation structure by the use of variograms and unconditional simulations, and obtain an effective s ample s ize, w hich, w hen u sed a s t he s ample s ize for t he bootstrapped samples for bagging, allows us to treat the data as independent. We test our hypothesis by assessing its prediction accuracy, and uncertainty quantification goodness, by running it on an extensive collection of synthetic datasets of varying correlation structures and noise levels, and a case study coming from a real dataset. We show that this new method is robust against overfitting and p erforms substantially better than standard bagging for spatial data. Our method has equal prediction accuracy to standard bagging when there is no noise in the data, and no problem of overfitting, and our model yields b etter prediction accuracy than standard bagging  \nfor increasing levels of noise in data, demonstrating our model’s superiority where the data is noisy. Our method yields substantially better uncertainty modeling capabilities compared to standard bagging for all noise levels, and this superiority increases for increasing levels of noise. We believe that our proposed method can be used in any case of n-dimensional spatial data, and it shall yield a significant advantage compared to standard bagging.  \nTable of Contents  \nList of Figures ................................... 8  \nChapter 1: Introduction ............................. 11  \nChapter 2: Spatial Bagging for Prediction ................... 14  \n2.1 Introduction ................................ 14  \n2.2 Methodology ................................ 21  \n2.3 Results ................................... 23  \nChapter 3: Spatial Bagging for Uncertainty Quantification .......... 32  \n3.1 Introduction ................................ 32  \n3.1.1 Uncertainty Goodness ....................... 35  \n3.2 Methodology ................................ 38  \n3.3 Results ................................... 39  \nChapter 4: Case Study .............................. 48  \n4.1 Introduction and Data Description .................... 48  \n4.2 Results ................................... 48  \n4.2.1 Case 1 ................................ 50  \n4.2.2 Case 2 ................................ 51  \n4.2.3 Case 3 ................................ 51  \n4.2.4 Case 4 ................................ 52  \nChapter 5: Conclusion .............................. 54  \nWorks Cited ..................................... 56  \nList of Figures  \n2.1 An example of a generated dataset, characterized by major variogram ","cbCaivhjh9l6byaP","https://ap.wps.com/l/cbCaivhjh9l6byaP","pdf",40344368,1,60,"English","en",105,"# List of Figures\n# Chapter 1: Introduction\n# Chapter 2: Spatial Bagging for Prediction\n## 2.1 Introduction\n## 2.2 Methodology\n## 2.3 Results\n# Chapter 3: Spatial Bagging for Uncertainty Quantification\n## 3.1 Introduction\n## 3.1.1 Uncertainty Goodness\n## 3.2 Methodology\n## 3.3 Results\n# Chapter 4: Case Study\n## 4.1 Introduction and Data Description\n## 4.2 Results\n## 4.2.1 Case 1\n## 4.2.2 Case 2\n## 4.2.3 Case 3\n## 4.2.4 Case 4\n# Chapter 5: Conclusion\n# Works Cited","[{\"question\":\"What problem does the thesis address in predictive ensemble machine learning?\",\"answer\":\"The thesis targets statistical inference and prediction when data are spatially correlated, where standard machine learning assumptions of independence do not hold.\"},{\"question\":\"How does the proposed spatial bagging method handle spatial dependence?\",\"answer\":\"It captures spatial correlation using variograms and unconditional simulations, then computes an effective sample size to make bootstrapped samples behave as independent.\"},{\"question\":\"How does the method perform compared with standard bagging, especially under noise?\",\"answer\":\"It is robust against overfitting and delivers substantially better uncertainty quantification for all noise levels, with equal prediction accuracy in noise-free data and improved prediction accuracy as noise increases.\"}]","A Novel Spatial Bagging Algorithm for Predictive Ensemble Machine Learning - Master of Science Thesis | PDF",1785808562,151,{"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},"a-novel-spatial-bagging-algorithm-for-predictive-ensemble-machine-learning-master-of-science-thesis","",{"@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/a-novel-spatial-bagging-algorithm-for-predictive-ensemble-machine-learning-master-of-science-thesis/122049/",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","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"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 predictive ensemble machine learning?","Question",{"text":75,"@type":76},"The thesis targets statistical inference and prediction when data are spatially correlated, where standard machine learning assumptions of independence do not hold.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed spatial bagging method handle spatial dependence?",{"text":80,"@type":76},"It captures spatial correlation using variograms and unconditional simulations, then computes an effective sample size to make bootstrapped samples behave as independent.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the method perform compared with standard bagging, especially under noise?",{"text":84,"@type":76},"It is robust against overfitting and delivers substantially better uncertainty quantification for all noise levels, with equal prediction accuracy in noise-free data and improved prediction accuracy as noise increases.","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},[93,97,101,105,109,114,119,122,127,130,134],{"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":21,"slug":108},5,"Comic","comic",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]