[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127551-en":3,"doc-seo-127551-105":31,"detail-sidebar-cat-0-en-105":92},{"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},127551,687207017582,"Himbo","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","A Versatile Federated Machine Learning Strategy with Applications to XGBoost, GMM and DBSCAN - Thesis Abstract","Machine learning advances have enabled solutions to complex problems, yet many approaches rely on centralized training where data and software reside on the same computing entity. For privacy-critical use cases such as biomedical informatics and user analytics, regulations can restrict moving personal information outside defined data silos. Federated machine learning addresses this by training partial models locally and aggregating them on a server without sharing user data. This thesis proposes a federated XGBoost implementation, introduces “FIN” as a Flower strategy, and presents a federated clustering method combining DBSCAN and Gaussian Mixture Models, evaluating efficiency on real and simulated data.","A Versatile Federated Machine Learning Strategy with Applications to XGBoost, GMM  \nand DBSCAN  \nIoanna Vasilopoulou  \nThesis submitted in partial fulfillment of the requirements for the Masters’ of Science degree in Bioinformatics  \nUniversity of Crete  \nSchool of Medicine  \nVoutes University Campus, 700 13 Heraklion, Crete, Greece  \nThesis Advisors:  \nDr. George Potamias,  \nDr. Alexandros Kanterakis  \nThis work has been performed at the University of Crete, School of Medicine.  \nThe work has been supported by the Foundation for Research and Technology - Hellas (FORTH), Institute of Computer Science (ICS) .  \nUniversity of Crete  \nSchool of Medicine  \nA Versatile Federated Machine Learning Strategy with Applications to  \nXGBoost, GMM and DBSCAN  \nThesis submitted by  \nIoanna Vasilopoulou  \nin partial fulfillment of the requirements for the Masters’ of Science degree in Bioinformatics  \nAuthor: Ioanna Vasilopoulou  \nThree member committee: George Potamias  \nThesis Supervisor  \nAlexandros Kanterakis  \nCommittee Member  \nPavlos Pavlidhs  \nCommittee Member  \nHeraklion, February 2023  \nA Versatile Federated Machine Learning Strategy with Applications to XGBoost, GMM and DBSCAN  \nAbstract  \nLatest algorithmic, hardware and programming advancements have brought Machine Learning (ML) closer to the solution of increasingly more complex problems. However, most of these solutions are only applicable to centralized approaches, where the complete set of training data and the analysis software are located on the same computational entity. For privacy critical applications like biomedical informatics and user data analytics this can be an important issue. Simply, very strict and active legislations on privacy and security, forbid the transfer of any personal information outside of a well defined data sylo (i.e. hospital, mobile device) .  \nFederated Machine Learning (FML) comes to address this issue by offering a different learning approach. Through an iterative process, partial models are trained in each data-sylo and then transferred into a server and aggregated into a single model without exchanging user data. One of the most known frameworks for FMLis Flower which deals with all trivial tasks like client-server communication with consistency and security. Flower offers many strategies with which ML methods can coordinate through the process of local training and model aggregation. Existing strategies focus mainly on simple ML methods mainly in the area of statistical learning (i.e. regression) and deep neural networks where model aggregation is straight forward. For more advanced ML methods like tree-based for classification, and distance-based for clustering, not only there is no Flower strategy, but even the concept of “federalization” is an active field of research.  \nHere we first create a new “Federalized” implementation of the well known tree-based XGBoost algorithm for classification. Then we introduce “FIN”, a novel Flower strategy that allows the inclusion of more sophisticated ML methods and we demonstrate how it can be used in our XGBoost implementation. Finally we implement a novel Federated data clustering method based on the synergy of DBSCAN and Gaussian Mixture Models. For all our implementations we perform an extensive set of tests in real and in simulated data and we demonstrate how these federated implementations have the same efficiency as their “classic” counterparts.  \nAcknowledgements  \nFirst of all, I would like to thank my supervisor, Dr. Potamias, for his valuable guidance. I also want to express my deepest gratitude to my advisor, Dr. Alexandros Kanterakis, for his wise advice, support, and encouragement throughout the entire research process. His advice have greatly improved the quality and clarity of my work that contributed to my academic and technical growth. I would also like to thank Alexander Shevtsov for his helpful discussions and suggestions. Finally, I would like to thank my family and friends for t","cbCainlsPTTSFGei","https://ap.wps.com/l/cbCainlsPTTSFGei","pdf",1535257,2,1,50,"English","en",105,"# Introduction\n## Federated Learning\n## Federated Frameworks\n## Federated Machine Learning algorithms\n## Purpose of the study\n# Datasets\n# Federated XGBoost\n## Preliminaries of XGBoost\n## XGBoost on a Federated Learning setting\n## Implementation\n## Experiments\n# FIN Strategy\n## Implementation of FIN\n# GMM and DBSCAN on a Federated Learning setting\n## Federated GMM\n## Experiments\n## Federated DBSCAN based on GMM data description\n# Discussion\n## Bibliography","[{\"question\":\"Why is federated machine learning needed for biomedical informatics and similar domains?\",\"answer\":\"Centralized ML requires data transfer to a single computing entity, which privacy and security regulations may prohibit. Federated learning keeps data inside defined silos while enabling model training and aggregation.\"},{\"question\":\"What is the purpose of the proposed federated XGBoost implementation?\",\"answer\":\"The thesis introduces a “Federalized” XGBoost approach for classification, enabling tree-based learning within a federated learning setup.\"},{\"question\":\"How do FIN and the DBSCAN–GMM federated clustering method contribute to the thesis?\",\"answer\":\"FIN provides a novel Flower strategy to incorporate more sophisticated ML methods into federated training, and the thesis develops a federated clustering method that synergizes DBSCAN and Gaussian Mixture Models.\"}]","A Versatile Federated Machine Learning Strategy with Applications to XGBoost, GMM and DBSCAN - Thesis Abstract | PDF",1785939915,126,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":29},"a-versatile-federated-machine-learning-strategy-with-applications-to-xgboost-gmm-and-dbscan-thesis-abstract","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/a-versatile-federated-machine-learning-strategy-with-applications-to-xgboost-gmm-and-dbscan-thesis-abstract/127551/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why is federated machine learning needed for biomedical informatics and similar domains?","Question",{"text":76,"@type":77},"Centralized ML requires data transfer to a single computing entity, which privacy and security regulations may prohibit. Federated learning keeps data inside defined silos while enabling model training and aggregation.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What is the purpose of the proposed federated XGBoost implementation?",{"text":81,"@type":77},"The thesis introduces a “Federalized” XGBoost approach for classification, enabling tree-based learning within a federated learning setup.",{"name":83,"@type":74,"acceptedAnswer":84},"How do FIN and the DBSCAN–GMM federated clustering method contribute to the thesis?",{"text":85,"@type":77},"FIN provides a novel Flower strategy to incorporate more sophisticated ML methods into federated training, and the thesis develops a federated clustering method that synergizes DBSCAN and Gaussian Mixture Models.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":22,"slug":114},6,"Technology","technology",{"id":116,"doc_module":4,"doc_module_name":47,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":47,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":47,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":47,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]