[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-189879-105":3,"detail-sidebar-cat-1-en-105":73,"doc-detail-189879-en":118},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":66,"head_meta":68,"extra_data":70,"updated_unix":72},105,"en","deep-neural-network-training-and-testing-performance-comparison","Deep Neural Network - Training and Testing Performance Comparison","","A performance comparison study of deep neural network architectures for source-to-target task transfer across varying dataset sizes. Experiments define source task (AV) and target task (S4), using train/test splits and structured embeddings. Results report multiple accuracy metrics (word accuracy, template word accuracy, F-score, and line accuracy) alongside processing time for training and testing. Configurations evaluate DTNN against hybrid CRF-DNN-DTNN variants to assess scalability and effectiveness under different numbers of logs/templates/clusters.",{"@graph":14,"@context":65},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/template/","Template",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/template/presentations/","Presentations",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/template/deep-neural-network-training-and-testing-performance-comparison/189879/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/deep-neural-network-training-and-testing-performance-comparison/189879.png","ImageObject",442,249,{"name":42,"@type":43},"Valentina","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-02","2026-09-03",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",5,{"@type":57,"mainEntity":58},"FAQPage",[59],{"name":60,"@type":61,"acceptedAnswer":62},"How do processing times relate to the model evaluations?","Question",{"text":63,"@type":64},"Processing time is provided for both training and testing, allowing comparison of accuracy changes alongside computational cost for different model configurations.","Answer","https://schema.org",{"og:url":32,"og:type":67,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":69,"canonical":32},"index,follow",{"doc_id":71,"site_id":7},189879,1788399767,{"code":4,"msg":74,"data":75},"success",[76,80,85,90,95,100,105,110,114],{"id":77,"doc_module":22,"doc_module_name":25,"category_name":29,"show_sort_weight":78,"slug":79},11,90,"presentations",{"id":81,"doc_module":22,"doc_module_name":25,"category_name":82,"show_sort_weight":83,"slug":84},12,"Resumes",80,"resumes",{"id":86,"doc_module":22,"doc_module_name":25,"category_name":87,"show_sort_weight":88,"slug":89},14,"Invoices",70,"invoices",{"id":91,"doc_module":22,"doc_module_name":25,"category_name":92,"show_sort_weight":93,"slug":94},15,"Posters",60,"posters",{"id":96,"doc_module":22,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},16,"Social Media",50,"social-media",{"id":101,"doc_module":22,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},17,"Forms",40,"forms",{"id":106,"doc_module":22,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},18,"Letters",30,"letters",{"id":111,"doc_module":22,"doc_module_name":25,"category_name":112,"show_sort_weight":55,"slug":113},21,"Paper Templates","papers-templates",{"id":115,"doc_module":22,"doc_module_name":25,"category_name":116,"show_sort_weight":4,"slug":117},158,"General","general-158",{"code":4,"msg":74,"data":119},{"doc_id":71,"user_id":120,"nickname":42,"user_avatar":121,"doc_module":22,"category_id":77,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":122,"file_id":123,"file_url":124,"file_type":125,"file_size":126,"view_count":127,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":128,"language":129,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":130,"faqs":131,"seo_title":132,"seo_description":12,"update_tm":72,"read_time":26},13056703020460,"https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923","| yyyy | on |  | pts/0 |\n| --- | --- | --- | --- |\n\n|  | Deep Neural Network (DNN) |  |\n| --- | --- | --- |\n|  Embedding \u003Cbr>\u003Cbr>Neural Network\u003Cbr>Char Network\u003Cbr>|  |  |\n\n\n|  | Source Task (AV) |  | Target Task (S4) |  |\n| --- | --- | --- | --- | --- |\n|  | Train (SX) | Test (SY) | Train (TX) | Test (TY) |\n| \\# Logs | 100000 | 100000 | 74496 | 74910 |\n| \\# Templates | 100000 | 100000 | 74496 | 74910 |\n| \\# Clusters | 38 | 45 | 41 | 75 |\n\n\n| \\# Train.\u003Cbr>Data | Methods | Accuracy Metrics |  |  |  | Processing time (sec) |  |\n| --- | --- | --- | --- | --- | --- | --- | --- |\n|  |  | Word Acc. | Template Word Acc. | F Acc. | Line Acc. | Training T. | Testing T. |\n| 0 | DTNN | 0.898 􀀆 8.73E-7 | 0.838 􀀆 5.45E-6 | 0.797 􀀆 1.07E-5 | 0.341 􀀆 7.04E-6 | 0.341 􀀆 7.04E-6 | 0.341 􀀆 7.04E-6 |\n| 1 | CRF\u003Cbr>DNN\u003Cbr>DTNN | 0.779 􀀆 0.025\u003Cbr>0.677 􀀆 0.064\u003Cbr>0.912 􀀆 0.02 | 0.772 􀀆 0.063\u003Cbr>0.603 􀀆 0.08\u003Cbr>0.871 􀀆 0.022 | 0.828 􀀆 0.115\u003Cbr>0.888 􀀆 0.11\u003Cbr>0.949 􀀆 0.088 | 0.228 􀀆 0.081\u003Cbr>0.163 􀀆 0.13\u003Cbr>0.442 􀀆 0.103 | 2E-7 􀀆 0.0008 27.80 􀀆 0.2099 1877.56 􀀆 3.996 | 4.31 􀀆 0.07\u003Cbr>82.28 􀀆 0.232\u003Cbr>171.34 􀀆 9.362 |\n| 10 | CRF\u003Cbr>DNN\u003Cbr>DTNN | 0.932 􀀆 0.025\u003Cbr>0.958 􀀆 0.031\u003Cbr>0.967 􀀆 0.028 | 0.818 􀀆 0.067\u003Cbr>0.804 􀀆 0.03\u003Cbr>0.869 􀀆 0.033 | 0.991 􀀆 0.005\u003Cbr>0.956 􀀆 0.045\u003Cbr>0.950 􀀆 0.057 | 0.786 􀀆 0.027\u003Cbr>0.801 􀀆 0.101\u003Cbr>0.858 􀀆 0.111 | 0.01 􀀆 0.0004\u003Cbr>31.47 􀀆 0.1487\u003Cbr>2169.65 􀀆 3.780 | 4.3 􀀆 0.093\u003Cbr>82.69 􀀆 0.452\u003Cbr>185.88 􀀆 1.455 |\n| 100 | CRF\u003Cbr>DNN\u003Cbr>DTNN | 0.990 􀀆 0.004\u003Cbr>0.994 􀀆 0.003\u003Cbr>0.996 􀀆 0.001 | 0.901 􀀆 0.04\u003Cbr>0.895 􀀆 0.014\u003Cbr>0.926 􀀆 0.007 | 0.998 􀀆 0.001\u003Cbr>0.999 􀀆 0.0002\u003Cbr>0.993 􀀆 0.004 | 0.940 􀀆 0.021\u003Cbr>0.976 􀀆 0.011\u003Cbr>0.972 􀀆 0.011 | 0.02 􀀆 0.0029\u003Cbr>74.74 􀀆 0.0955\u003Cbr>2305.77 􀀆 11.915 | 4.39 􀀆 0.114\u003Cbr>82.80 􀀆 0.462\u003Cbr>186.39 􀀆 0.891 |\n| 1000 | CRF\u003Cbr>DNN\u003Cbr>DTNN | 0.999 􀀆 0.0003\u003Cbr>0.998 􀀆 0.0002\u003Cbr>0.999 􀀆 0.0003 | 0.933 􀀆 0.002\u003Cbr>0.921 􀀆 0.009\u003Cbr>0.937 􀀆 0.003 | 1.000 􀀆 5.4E-05 0.999 􀀆 0.0002 1.000 􀀆 0.0003 | 0.993 􀀆 0.001\u003Cbr>0.989 􀀆 0.002\u003Cbr>0.993 􀀆 0.0016 | 0.08 􀀆 0.028\u003Cbr>74.70 􀀆 0.091\u003Cbr>2704.20 􀀆 8.487 | 4.26 􀀆 0.084\u003Cbr>82.74 􀀆 0.29\u003Cbr>186.17 􀀆 0.591 |\n| 10000 | CRF\u003Cbr>DNN\u003Cbr>DTNN | 0.999 􀀆 4.36E-5\u003Cbr>0.998 􀀆 0.0003\u003Cbr>1.000 􀀆 1.52E-5 | 0.943 􀀆 0.002\u003Cbr>0.927 􀀆 0.007\u003Cbr>0.961 􀀆 0.002 | 1.000 􀀆 5. 1E-9\u003Cbr>1.000 􀀆 0.0003\u003Cbr>1.000 􀀆 1.32E-5 | 0.996 􀀆 0.0003\u003Cbr>0.989 􀀆 0.0017\u003Cbr>0.998 􀀆 9.3E-5 | 4.32 􀀆 1.2473\u003Cbr>132.66 􀀆 0.058\u003Cbr>4032.86 􀀆 7.79 | 4.42 􀀆 0.088\u003Cbr>83.11 􀀆 0.258\u003Cbr>185.42 􀀆 0.884 |","cbCaio4OA4lh1zoF","https://ap.wps.com/l/cbCaio4OA4lh1zoF","pdf",376872,8,6,"English","# Experimental Setup\n## Source and Target Tasks\n## Models and Embedding Network\n## Dataset Partitions and Scales\n# Results and Metrics\n## Accuracy Metrics\n## Processing Time (Training/Testing)","[{\"question\":\"How do processing times relate to the model evaluations?\",\"answer\":\"Processing time is provided for both training and testing, allowing comparison of accuracy changes alongside computational cost for different model configurations.\"}]","Deep Neural Network - Training and Testing Performance Comparison | PDF"]