[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-272087-105":59,"doc-detail-272087-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","leveraging-morphological-and-lexical-features-in-synthetic-data-generation-for-dialect-specific-machine-translation-thesis","Leveraging Morphological and Lexical Features in Synthetic Data Generation for Dialect-Specific Machine Translation - Thesis","","This thesis investigates language-agnostic strategies to improve neural machine translation for low-resource language variants by leveraging morphological and lexical information. Linguistic features are extracted to create synthetic text data guided by a standard or more dominant language variant, with emphasis on German dialect groups Bavarian and Alemannic. The work details a methodology and data pipeline, defines an evaluation framework, and reports results through baseline comparisons and perturbation-based translation experiments, followed by discussion of limitations and future directions.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/leveraging-morphological-and-lexical-features-in-synthetic-data-generation-for-dialect-specific-machine-translation-thesis/272087/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/leveraging-morphological-and-lexical-features-in-synthetic-data-generation-for-dialect-specific-machine-translation-thesis/272087.png","ImageObject",300,407,{"name":92,"@type":93},"Ben Jamin","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-24","2026-09-15",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What is the main goal of the thesis?","Question",{"text":112,"@type":113},"To improve neural machine translation for low-resource language variants using language-agnostic methods based on morphological and lexical feature extraction for synthetic data generation.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"Which German dialect groups are studied?",{"text":117,"@type":113},"The thesis focuses on the Bavarian and Alemannic dialect groups.",{"name":119,"@type":110,"acceptedAnswer":120},"How are improvements evaluated?",{"text":121,"@type":113},"Results are assessed by comparing a German-English baseline with perturbation-based machine translation, supported by both quantitative and qualitative analysis frameworks.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},272087,1789463125,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":144,"preview_page_count":4},2336478466772,"https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2","UNIVERSITY OF HAMBURG  \nMASTERS THESIS  \nLeveraging Morphological and Lexical Features in Synthetic Data Generation for Dialect-Specific  \nMachine Translation  \nAuthor Christian Schuler  \n(6449321)  \nSupervisor Dr. Sina Ahmadi  \nUniversity of Zurich  \nSupervisor Dr. Seid Muhie Yimam University of Hamburg  \nExaminer Prof. Dr. Chris Biemann University of Hamburg  \nA thesis submitted in fulfillment of the requirements for the Master of Science Informatik  \nof the  \nLanguage Technology Faculty of Informatics Fachbereich Informatik  \nJuly, 2024  \ni  \n“ Wege entstehen dadurch, dass man sie geht.”  \n(Paths are made by walking them.)  \nFranz Kafka  \nii  \nAcknowledgements  \nI am deeply grateful to my thesis supervisors Dr. Seid Muhie Yimam and Dr. Sina Ahmadi. Their meticulous review of drafts and invaluable feedback significantly helped me to improve the style and coherence of this thesis. Their goal-oriented guidance was instrumental in keeping me focused on the core objectives, helping me navigate the everexpanding labyrinth of intriguing side topics. It was a particular honor to work with Sina on these topics; I’m especially thankful for his willingness to supervise me externally and across significant time zones, first from Virginia and later from Zurich. I would like to extend my appreciation to Prof. Dr. Chris Biemann, my thesis examiner, for his time and expertise in evaluating this work and providing the opportunity for me to write this thesis in the Language Technology Group at University of Hamburg.  \nI have to thank Raman, who, even if by mere happenstance (during a conversation in our student-dormitory’s shared kitchen), aimed me like an arrow towards the topic of my thesis. I would also like to express my heartfelt gratitude to Anran and Domi, whose companionship and support made this journey not just possible, but enjoyable. Their assistance extended beyond emotional support to practical help, exemplified by Domi driving me and all my belongings across the country on the weekend of the thesis submission-a gesture that truly encapsulates the true spirit of love.  \niii  \nUNIVERSITY OF HAMBURG  \nAbstract  \nFaculty of Informatics  \nFachbereich Informatik  \nMaster of Science Informatik  \nLeveraging Morphological and Lexical Features in Synthetic Data Generation for Dialect-Specific Machine Translation  \nby Christian Schuler  \nThis work aims to explore language-agnostic methods for improving neural machine translation of low-resource language variants. Extracting linguistic features enables the creation of synthetic text data oriented by a standard or more dominant language variant. In the main focus are the German dialect groups Bavarian and Alemannic.  \nKeywords: Natural language processing, dialectal varieties, linguistic information, low-resourced language, neural machine translation  \niv  \nContents  \nAcknowledgements ii  \nAbstract iii  \nContents iv  \nList of Figures vi  \nList of Tables vii  \nList of Abbreviations and Acronyms viii  \n1 Introduction 1  \n1.1 Natural Language Processing and Linguistics ................ 1  \n1.2 Low-Resource Languages . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2  \n1.3 Machine Translation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3  \n1.4 Objectives and Research Questions . . . . . . . . . . . . . . . . . . . . . . 3  \n1.5 Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4  \n2 Motivation 5  \n2.1 Languages and Dialects ............................ 5  \n2.1. 1 Dialect Continua . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7  \n2.1.2 Mutual Understanding between Dialects . . . . . . . . . . . . . . . 8  \n2.2 The Notion of Low-Resource . . . . . . . . . . . . . . . . . . . . . . . . . 9  \n2.3 Language Classification Issues . . . . . . . . . . . . . . . . . . . . . . . . . 11  \n2.4 Machine Translation and Dialects . . . . . . . . . . . . . . . . . . . . . . . 12  \n3 Related Work 14  \n3.1 NLP building on Linguistics ...............","cbCaijyIHMQRVJj0","https://ap.wps.com/l/cbCaijyIHMQRVJj0","pdf",5174667,87,"English","# Acknowledgements\n# Abstract\n# Contents\n# List of Figures\n# List of Tables\n# List of Abbreviations and Acronyms\n# 1 Introduction\n## 1.1 Natural Language Processing and Linguistics\n## 1.2 Low-Resource Languages\n## 1.3 Machine Translation\n## 1.4 Objectives and Research Questions\n## 1.5 Outline\n# 2 Motivation\n## 2.1 Languages and Dialects\n## 2.2 The Notion of Low-Resource\n## 2.3 Language Classification Issues\n## 2.4 Machine Translation and Dialects\n# 3 Related Work\n## 3.1 NLP building on Linguistics\n## 3.2 Machine Translation\n## 3.3 Low-Resource Machine Translation Evaluation\n## 3.4 Approaches to Low-Resource Machine Translation\n## 3.5 Synthetic Text Data Generation\n## 3.6 Bilingual Lexicon Induction\n## 3.7 Linguistic Dialectal Features for Translation\n# 4 Methodology and Data\n## 4.1 Proposed Methodology\n## 4.2 An Approach with Multiple Angles\n## 4.3 Utilizing Data from DialectBLI\n# 5 Results and Evaluation\n## 5.1 Created Perturbations\n## 5.2 Machine Translation Baseline for German-English\n## 5.3 Perturbation-Based Machine Translation\n# 6 Discussion and Conclusion\n## 6.1 Interpretation of the Results\n## 6.2 Ongoing Challenges and Future Work","[{\"question\":\"What is the main goal of the thesis?\",\"answer\":\"To improve neural machine translation for low-resource language variants using language-agnostic methods based on morphological and lexical feature extraction for synthetic data generation.\"},{\"question\":\"Which German dialect groups are studied?\",\"answer\":\"The thesis focuses on the Bavarian and Alemannic dialect groups.\"},{\"question\":\"How are improvements evaluated?\",\"answer\":\"Results are assessed by comparing a German-English baseline with perturbation-based machine translation, supported by both quantitative and qualitative analysis frameworks.\"}]","Leveraging Morphological and Lexical Features in Synthetic Data Generation for Dialect-Specific Machine Translation - Thesis | PDF",219]