[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127670-en":3,"doc-seo-127670-105":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},127670,962084925636,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","PHYSICS-GUIDED MACHINE LEARNING FOR SMALL DATA SETS - Dissertation","Physics-guided machine learning for small data sets develops methods to support predictive maintenance in the bottling industry, aiming to prevent costly machine breakdowns and improve decision quality with limited labeled data. The work addresses challenges in machine error modeling, data acquisition, synchronization, and high-resolution preprocessing via statistical methods. It formulates physics-based bottle transfer measures, then designs semi-supervised and one-shot anomaly detection workflows, including feature extraction and evaluation using iNAB. Further sections propose physics-and expert-driven error sketch recognition with dynamic time warping and expert feedback.","PHYSICS-GUIDED MACHINE LEARNING FOR SMALL DATA SETS  \nDISSERTATION ZUR ERLANGUNG DES DOKTORGRADES DER NATURWISSENSCHAFTEN (DR. RER. NAT.) DER FAKULTÄT PHYSIK DER UNIVERSITÄT REGENSBURG  \nvorgelegt von Andrea Spichtinger aus  \nOberviechtach  \nim Jahr 2021  \nPromotionsgesuch eingereicht am:  \nDie Arbeit wurde angeleitet von: Prof. Dr. Elmar Lang  \nPrüfungsausschuss:  \nVorsitzender: Prof. Dr. Dieter Weiss  \n1. Gutachter: Prof. Dr. Elmar Lang  \n2. Gutachter: Dr. Stefan Solbrig Weiterer Prüfer: Prof. Dr. Klaus Richter  \nDatum Promotionskolloquium: 16.11.2021  \nTo all the people,  \nwho believed in me and supported me on my journey.  \nAnd to the ones,  \nwho will follow.  \nIn particular my godchild, who will be born, when I submit this thesis.  \nContents  \nAbstract 1  \n1 Challenges of Machine Learning in the Bottling Industry 3  \n2 Use Case and Data Processing 7  \n2.1 Analysis of most frequent Machine Errors .............. 7  \n2.2 Synchronization Error ......................... 9  \n2.3 Data Acquisition ............................ 10  \n2.4 Data Preprocessing: High Resolution via Statistics ......... 10  \n2.5 Model Deployment ........................... 12  \n2.6 Hardware and Software environment ................. 12  \n3 Physics of Bottle Transfer 15  \n3.1 Conservation measure: Position-based Velocity ............ 15  \n3.2 Bottle Transport ............................. 16  \n3.3 Bottle Handover ............................ 19  \n3.3.1 General Behavior ........................ 19  \n3.3.2 Bottle Handover without Friction ............... 19  \n3.3.3 Bottle Handover with Friction ................. 20  \n3.4 Faulty Handover (mathematical description) ............. 21  \n3.4.1 Introduction .......................... 21  \n3.4.2 System Modeling ........................ 21  \n3.4.3 Solution by Lagrangian Formalism .............. 23  \n3.4.4 Comparison Model Results with Reality ........... 26  \n3.4.5 Potential Model Enhancements ................ 33  \n3.5 Summary ................................ 34  \n4 Semi-supervised Anomaly Detection-Methods 37  \n4.1 Definition of Anomaly Detection .................... 37  \n4.2 Architecture of Semi-supervised Anomaly Detection ........ 40  \n4.3 Feature Extraction ........................... 43  \n4.3.1 Frequency based Features ................... 44  \n4.3.2 Blind Source Separation-Non-negative Matrix Factorization (NMF) ............................. 47  \n4.3.3 Massive Feature Extraction .................. 49  \n4.4 Semi-supervised Anomaly Measures .................. 50  \n4.4.1 Statistical Measures ...................... 51  \n4.4.2 Classification Measures .................... 53  \n4.4.3 Clustering Measures ...................... 54  \n4.4.4 Nearest Neighbor Measures .................. 56  \n4.4.5 Spectral Measures ....................... 57  \n4.4.6 Information theoretic Measures ................ 58  \n4.4.7 Collective Anomaly Measures ................. 59  \n4.5 Anomaly Evaluation .......................... 59  \n4.6 Improvements Methods ........................ 62  \n4.6.1 EMD for Semi-supervised Anomaly Detection ........ 62  \n4.6.2 Improve NAB score for real-world Labels (iNAB) ...... 65  \n4.7 Summary ................................ 68  \n5 One-shot Semi-supervised Anomaly Detection-Study 71  \n5.1 Structure Study ............................. 71  \n5.2 Details about Data and Labels ..................... 74  \n5.3 Modeling: One-shot Semi-supervised Anomaly Detection ..... 74  \n5.3.1 Label Healthy States ...................... 74  \n5.3.2 Feature Selection ........................ 75  \n5.3.3 One-shot Semi-supervised Anomaly Measure ........ 77  \n5.4 Evaluation iNAB ............................ 79  \n5.4.1 Error case preparation NAB .................. 79  \n5.4.2 Error case preparation iNAB .................. 82  \n5.4.3 Comparison NAB and iNAB .................. 83  \n5.5 Evaluation Anomaly Algorithms .................... 86  \n5.5.1 Evaluation Scores ........................ 86  \n5.5.2 Evaluation ........................... 90  \n5.6 C","cbCaiiM8fQCrcQmJ","https://ap.wps.com/l/cbCaiiM8fQCrcQmJ","pdf",15522381,2,1,173,"English","en",105,"# Abstract\n# Challenges of Machine Learning in the Bottling Industry\n# Use Case and Data Processing\n## Analysis of most frequent Machine Errors\n## Synchronization Error\n## Data Acquisition\n## Data Preprocessing: High Resolution via Statistics\n## Model Deployment\n## Hardware and Software environment\n# Physics of Bottle Transfer\n## Conservation measure: Position-based Velocity\n## Bottle Transport\n## Bottle Handover\n## Faulty Handover (mathematical description)\n# Semi-supervised Anomaly Detection-Methods\n## Definition of Anomaly Detection\n## Architecture of Semi-supervised Anomaly Detection\n## Feature Extraction\n## Semi-supervised Anomaly Measures\n## Anomaly Evaluation\n## Improvements Methods\n# One-shot Semi-supervised Anomaly Detection-Study\n## Structure Study\n## Details about Data and Labels\n## Modeling: One-shot Semi-supervised Anomaly Detection\n## Evaluation iNAB\n## Evaluation Anomaly Algorithms\n## Compare Winning Method with Physical Method\n## Transfer Learning to other Bottlers\n# Physics-and Expert-Driven Error Sketch Recognition\n## Literature Research\n## Sketch Preparation\n## Study Setup\n## Results\n## Feedback and Retraining\n## Optimize Scoring Time\n# Conclusion and Outlook","[{\"question\":\"What problem does this dissertation target in the bottling industry?\",\"answer\":\"It targets costly machine breakdowns by enabling proactive predictive maintenance using machine learning under small-data constraints.\"},{\"question\":\"How is the dissertation's physics guidance incorporated?\",\"answer\":\"It introduces physics-based measures for bottle transfer and integrates physics constraints into modeling and error descriptions, such as position-based velocity and faulty handover formulations.\"},{\"question\":\"Which anomaly detection approaches are proposed for limited labels?\",\"answer\":\"The work presents semi-supervised anomaly detection methods and a one-shot semi-supervised study, supported by feature extraction, anomaly measures, and evaluation using NAB and iNAB.\"}]","PHYSICS-GUIDED MACHINE LEARNING FOR SMALL DATA SETS - Dissertation | PDF",1785940646,436,{"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},"physics-guided-machine-learning-for-small-data-sets-dissertation","",{"@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/physics-guided-machine-learning-for-small-data-sets-dissertation/127670/",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},"What problem does this dissertation target in the bottling industry?","Question",{"text":76,"@type":77},"It targets costly machine breakdowns by enabling proactive predictive maintenance using machine learning under small-data constraints.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How is the dissertation's physics guidance incorporated?",{"text":81,"@type":77},"It introduces physics-based measures for bottle transfer and integrates physics constraints into modeling and error descriptions, such as position-based velocity and faulty handover formulations.",{"name":83,"@type":74,"acceptedAnswer":84},"Which anomaly detection approaches are proposed for limited labels?",{"text":85,"@type":77},"The work presents semi-supervised anomaly detection methods and a one-shot semi-supervised study, supported by feature extraction, anomaly measures, and evaluation using NAB and iNAB.","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},[94,98,102,106,111,116,121,124,129,132,136],{"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":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":107,"slug":139},19,"General","general"]