[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126352-en":3,"doc-seo-126352-105":31,"detail-sidebar-cat-0-en-105":93},{"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126352,962085564381,"Clementine","https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8",8,"Research & Report","Predicting Stellar Ages from Chemical Abundances with XGBoost - A Machine Learning Approach to Galactic Archaeology - Master of Science Thesis","Master of Science thesis investigating how stellar ages can be predicted from chemical abundance patterns using eXtreme Gradient Boosting (XGBoost) in the context of galactic archaeology. The work integrates stellar-age inference through Bayesian isochrone fitting, then builds and tunes an XGBoost model trained on GALAH DR4 data with dedicated training/test selection. Model quality is assessed via cross-validation, RMSE on training and test sets, and residual and uncertainty behavior, complemented by feature-importance analysis and age–abundance relationships.","UNIVERSITY OF OKLAHOMA  \nGRADUATE COLLEGE  \nPREDICTING STELLAR AGES FROM CHEMICAL ABUNDANCES WITH  \nXGBOOST:  \nA MACHINE LEARNING APPROACH TO GALACTIC ARCHAEOLOGY  \nA THESIS  \nSUBMITTED TO THE GRADUATE FACULTY in partial fulfillment of the requirements for the Degree of  \nMASTER OF SCIENCE  \nBy TUCKER CAPPS  \nNorman, Oklahoma  \nPREDICTING STELLAR AGES FROM CHEMICAL ABUNDANCES WITH  \nXGBOOST:  \nA MACHINE LEARNING APPROACH TO GALACTIC ARCHAEOLOGY  \nA THESIS APPROVED FOR THE  \nGALLOGLY COLLEGE OF ENGINEERING  \nBY THE COMMITTEE CONSISTING OF  \nDr. Charles Nicholson, Chair  \nDr. Andres Gonzalez Huertas  \nDr. Michael Hayden  \n©Copyright by TUCKER CAPPS 2025 All Rights Reserved.  \nContents  \nList of Tables v  \nList of Figures vi  \nAbstract viii  \n1 Introduction 1  \n1. 1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1  \n1.2 Objectives ................................. 2  \n1.3 Scope and Significance ......................... 3  \n2 Background 4  \n2.1 Stellar Ages and Isochrone Fitting ................... 4  \n2.2 Chemical Abundances as Age Indicators ................ 8  \n2.3 Machine Learning for Stellar Age Determination ........... 10  \n3 Data 12  \n3.1 GALAH Data Release 4 ......................... 12  \n3.2 Data Selection .............................. 13  \n3.3 Training+Test Set ............................ 18  \n4 Methodology 19  \n4.1 Bayesian Isochrone Fitting ....................... 19  \n4.2 eXtreme Gradient Boosting (XGBoost) ................ 22  \n5 Results 28  \n5.1 Cross-Validation and Final Hyperparamters ............. 28  \n5.2 Model Performance on Training Set .................. 29  \n5.3 Model Performance on Test Set ..................... 30  \n5.4 Residual Analysis ............................ 33  \n5.4.1 Training Set Residuals ..................... 33  \n5.4.2 Test Set Residuals ........................ 38  \n5.4.3 Test Set Residuals vs Chemical Abundances ......... 41  \n5.5 Feature Importance ........................... 43  \n5.6 Age-Abundance Relations ........................ 45  \n6 Discussion 51  \n7 Conclusion 56  \nReferences 58  \nList of Tables  \n1 Quality cuts used to select data from GALAH DR4 ........... 17  \n2 Additional cuts used to select MSTO data used for model training.. 19  \n3 Hyperparameter grid used to tune the XGBoost model, consisting of  \n103,680 possible hyperparameter combinations............ 28  \n4 Optimal configuration of hyperparameters that minimized the average RMSE across all folds......................... 29  \nList of Figures  \n1 “In the Hertzsprung-Russell diagram the temperatures of stars are plotted against their luminosities. The position of a star in the diagram provides information about its present stage and its mass. Stars that burn hydrogen into helium lie on the diagonal Branch, the  \nso-called Main Sequence.” Credit: ESO . . . . . . . . . . . . . . . 6  \n2 A model isochrone with solar metallicity and solar age. The near horizontal line of points near the bottom of the plot denotes the Main Sequence. The region in the red box is the Main Sequence TurnOff. The diagonal line that extends up and to the right is the Giant Branch. Each data point on this plot represents a star with a unique initial mass, with initial mass increasing as you move along the Main  \nSequence and up to the Giant Branch.................. 7  \n3 Periodic table of elements displaying the relative abundance from each nucleosynthetic production site. Credit: Kobayashi et al. (2020) 8  \n4 A set of theoretical isochrones having the same metallicity. Notice  \nhow the Main Sequence and Giant Branch are stacked on top of eachother while the data points in the Main Sequence Turn-Off are discernible................................... 11  \n5 A Kiel Diagram of 663,075 stars having flag_sp = 0 from the GALAH DR4 dataset. Notice the Main Sequence, Main Sequence Turn-Off, and Giant Branch............................. 14  \n6 A Kiel Diagram of 421,934 stars having good measurements for the desired elemental abunda","cbCaiaAyZr2rB5Gn","https://ap.wps.com/l/cbCaiaAyZr2rB5Gn","pdf",2773020,5,1,72,"English","en",105,"# Introduction\n## Motivation\n## Objectives\n## Scope and Significance\n# Background\n## Stellar Ages and Isochrone Fitting\n## Chemical Abundances as Age Indicators\n## Machine Learning for Stellar Age Determination\n# Data\n## GALAH Data Release 4\n## Data Selection\n## Training+Test Set\n# Methodology\n## Bayesian Isochrone Fitting\n## eXtreme Gradient Boosting (XGBoost)\n# Results\n## Cross-Validation and Final Hyperparamters\n## Model Performance on Training Set\n## Model Performance on Test Set\n## Residual Analysis\n## Feature Importance\n## Age-Abundance Relations\n# Discussion\n# Conclusion\n# References","[{\"question\":\"What is the core goal of the thesis?\",\"answer\":\"To predict stellar ages from chemical abundances using an XGBoost machine learning model for applications in galactic archaeology.\"},{\"question\":\"How are stellar ages and training targets obtained?\",\"answer\":\"Stellar ages are derived using Bayesian isochrone fitting, which provides ground-truth ages for model training and evaluation.\"},{\"question\":\"How is the XGBoost model evaluated and validated?\",\"answer\":\"The thesis reports cross-validation with hyperparameter tuning, then measures model performance using RMSE on both training and test sets, followed by residual analyses and feature-importance assessment.\"}]","Predicting Stellar Ages from Chemical Abundances with XGBoost - A Machine Learning Approach to Galactic Archaeology - Master of Science Thesis | PDF",1785904626,181,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"predicting-stellar-ages-from-chemical-abundances-with-xgboost-a-machine-learning-approach-to-galactic-archaeology-master-of-science-thesis","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/predicting-stellar-ages-from-chemical-abundances-with-xgboost-a-machine-learning-approach-to-galactic-archaeology-master-of-science-thesis/126352/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-22","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What is the core goal of the thesis?","Question",{"text":77,"@type":78},"To predict stellar ages from chemical abundances using an XGBoost machine learning model for applications in galactic archaeology.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How are stellar ages and training targets obtained?",{"text":82,"@type":78},"Stellar ages are derived using Bayesian isochrone fitting, which provides ground-truth ages for model training and evaluation.",{"name":84,"@type":75,"acceptedAnswer":85},"How is the XGBoost model evaluated and validated?",{"text":86,"@type":78},"The thesis reports cross-validation with hyperparameter tuning, then measures model performance using RMSE on both training and test sets, followed by residual analyses and feature-importance assessment.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":20,"slug":139},19,"General","general"]