[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127796-en":3,"doc-seo-127796-105":30,"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":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},127796,1099523885074,"Ivy","https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc",8,"Research & Report","Machine Learning Methods for Heterogeneous Data - Multimodal deep learning and dimensionality reduction - Academic dissertation","Data-driven methods transform modern applications by enabling predictive modeling from large-scale datasets. Real-world data appears in diverse formats, structures, and representations, and task concepts can be described ambiguously, creating mismatches in class and feature space distributions—a challenge known as data heterogeneity. In some settings, heterogeneity can be harnessed via multiple representations, but it reduces model robustness. This dissertation develops machine learning methods robust to heterogeneity or that exploit it, with attention to heterogeneous class distributions and dimensionality reduction.","KATERYNA CHUMACHENKO  \nMachine Learning Methods for Heterogeneous Data  \nMultimodal deep learning and dimensionality reduction  \nTampere University Dissertations 1091  \nTampere University Dissertations 1091  \nKATERYNA CHUMACHENKO  \nMachine Learning Methods for Heterogeneous Data Multimodal deep learning and dimensionality reduction  \nACADEMIC DISSERTATION  \nTo be presented, with the permission of the Faculty of Information Technology and Communication Sciences of Tampere University , for public discussion in the auditorium TB109 of the Tietotalo Building, Korkeakoulunkatu 1 , Tampere, on 4 October 2024 , at 12 o’clock.  \nACADEMIC DISSERTATION  \nTampere University, Faculty of Information Technology and Communication Sciences Finland  \nResponsible supervisor and Custos  \nSupervisor  \nPre-examiners  \nProfessor  \nMoncef Gabbouj Tampere University Finland  \nProfessor Alexandros Iosifidis Aarhus University Denmark  \nAssociate Professor Miguel Bordallo López University of Oulu Finland  \nProfessor  \nAbdullah Aydin Alatan  \nMiddle East Technical University Turkey  \nOpponent Professor  \nBenoit Macq  \nUniversité Catholique de Louvain  \nBelgium  \nThe originality of this thesis has been checked using the Turnitin OriginalityCheck service.  \nCopyright ©2024 author Cover design: Roihu Inc.  \nISBN 978-952-03-3600-4 (print)  \nISBN 978-952-03-3601-1 (pdf)  \nISSN 2489-9860 (print)  \nISSN 2490-0028 (pdf)  \n[http://urn.fi/URN:ISBN:978-952-03-3601-1](http://urn.fi/URN:ISBN:978-952-03-3601-1)  \nCarbon dioxide emissions from printing Tampere University dissertationshave been compensated.  \nPunaMusta Oy – Yliopistopaino Joensuu 2024  \nPREFACE/ACKNOWLEDGEMENTS  \nThe completion of this thesis would not have been possible without the encouragement and support of many wonderful people along the way.  \nFirst and foremost, I would like to express my sincere gratitude to my supervisors, Professor Moncef Gabbouj and Professor Alexandros Iosifidis, for their guidance, mentorship, and encouragement. Prof. Gabbouj has created a unique working environment, enabling me to explore multiple research areas and develop many valuable skills throughout my PhD journey. I am sincerely grateful for the opportunity to be a part of his research group. Prof. Iosifidis played a pivotal role in my decision to pursue a PhD, and has been a constant source of guidance and support throughout the process. I am grateful for the many insightful discussions we’ve had over the years, both technical and general. I am also indebted to Prof. Gabbouj and Prof. Iosifidis for their support towards my frequent detours from a conventional PhD path and for providing me with the freedom to explore different opportunities during these years.  \nI would also like to express my gratitude to Assistant Professor Jenni Raitoharju, who has been the first guide in my research journey during my Master’s thesis and has been supportive and encouraging to me and everyone in our lab.  \nDuring my PhD, I have been fortunate to have had the opportunity to pursue several internships in the industry. The number of inspiring people I have met during these times far exceeds what can be captured on this page, and I am deeply thankful to each and every one of them!  \nI have also been lucky to be surrounded by my amazing officemates and colleagues: Ali Senhaji, Anton Muravev, Aysen Degerli, Bilge Can Pullinen, Dat Tran Thanh, Fahad Sohrab, Farhad Pakdaman, Firas Laakom, IlkeAdalioglu, Mete Ahishali, Mohammad Al-Sa’d, Quoc Nguyen, SanazNami, Tuomas Jalonen, and others. Your support, fruitful coffee break discussions and gatherings outside of work have made this experience memorable and enjoyable. I wish you nothing but success in your  \nfuture endeavours!  \nI am forever grateful to my parents and my brother for supporting me throughout my life, and having trust in my decision to embark on my journey in Finland. Finally, the completion of this thesis would not have been possible without the unconditional love and support of my pa","cbCaib6zC31k7VDA","https://ap.wps.com/l/cbCaib6zC31k7VDA","pdf",10541906,1,194,"English","en",105,"# Preface/Acknowledgements\n# Abstract\n## Data heterogeneity and its impact\n## Exploiting heterogeneity vs improving robustness\n## Dissertation aims and focus","[{\"question\":\"What problem does the dissertation address?\",\"answer\":\"It addresses data heterogeneity, where class and feature space distributions differ due to multiple data representations and imprecise task-specific concepts.\"},{\"question\":\"How does the dissertation approach data heterogeneity?\",\"answer\":\"It develops methods that are either robust to heterogeneity or exploit it by using multiple representations for specific tasks.\"},{\"question\":\"What are the dissertation’s main focus areas?\",\"answer\":\"It focuses on heterogeneous data distributions with respect to class labels and studies dimensionality reduction methods that work under these conditions.\"}]","Machine Learning Methods for Heterogeneous Data - Multimodal deep learning and dimensionality reduction - Academic dissertation | PDF",1785941765,489,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-methods-for-heterogeneous-data-multimodal-deep-learning-and-dimensionality-reduction-academic-dissertation","",{"@graph":36,"@context":86},[37,54,69],{"@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/machine-learning-methods-for-heterogeneous-data-multimodal-deep-learning-and-dimensionality-reduction-academic-dissertation/127796/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-22","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},"What problem does the dissertation address?","Question",{"text":76,"@type":77},"It addresses data heterogeneity, where class and feature space distributions differ due to multiple data representations and imprecise task-specific concepts.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the dissertation approach data heterogeneity?",{"text":81,"@type":77},"It develops methods that are either robust to heterogeneity or exploit it by using multiple representations for specific tasks.",{"name":83,"@type":74,"acceptedAnswer":84},"What are the dissertation’s main focus areas?",{"text":85,"@type":77},"It focuses on heterogeneous data distributions with respect to class labels and studies dimensionality reduction methods that work under these conditions.","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":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]