[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120500-en":3,"doc-seo-120500-105":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},120500,13056703019662,"Evangeline","https://ap-avatar.wpscdn.com/avatar/be000253a8e92610077?_k=1778726343310543188",8,"Research & Report","Unlocking Viewer Insights in Linear Television - A Machine Learning Approach","Amid the shift toward digital television, linear TV advertising struggles with fragmented audiences, imprecise targeting, and rising costs. The study introduces a data-driven methodology that leverages extensive television first-party data together with machine learning models. Using state-of-the-art classification, the research achieves an average 88.7% accuracy for identifying household demographics. The resulting dataset and analytical pipeline support more precise advertiser decisions, enable personalized content delivery, and help reduce ad skipping.","Unlocking Viewer Insights in Linear Television: A Machine Learning Approach  \nJavier Carreno 1 , Khuong An Nguyen 1 , Zhiyuan Luo 1 , and Andrew Fish2  \n1 Royal Holloway University of London, Surrey TW20 0EX, United Kingdom  \n[javier.carreno.2023@live.rhul.ac.uk](javier.carreno.2023@live.rhul.ac.uk)  \n{Khuong.Nguyen, [Zhiyuan.Luo](Zhiyuan.Luo}@rhul.ac.uk)[}](Zhiyuan.Luo}@rhul.ac.uk)[@rhul.ac.uk](Zhiyuan.Luo}@rhul.ac.uk)  \n2 University of Liverpool, Liverpool L69 3BX, United Kingdom  \n[Andrew.Fish@liverpool.ac.uk](Andrew.Fish@liverpool.ac.uk)  \nAbstract. Amidst the TV digital transformation, traditional linear television advertising faces daunting challenges, including fragmented viewership, imprecise targeting, and increased cost. Therefore, this paper proposes a novel methodology, leveraging an extensive dataset of television first-party data and Machine Learning to understand the audience’s behaviour. By employing state-of-the-art machine learning classification models, we revealed an average 88.7% accuracy in correctly identifying the household demographics. This result offers promising outcomes for refining advertising strategies within linear television to enhance targeting precision, personalise content and mitigate ad skipping.  \nKeywords: Audience Insights · Machine Learning · Household Classification.  \n1 Introduction  \nLinear television represents the traditional approach to broadcasting, where TV networks adhere to predetermined schedules, airing specific content at scheduled times for viewers. Unlike modern on-demand or streaming services, linear TV restricts viewer control over content access, compelling them to tune in during scheduled broadcasts. Advertisers capitalise on this model by securing commercial slots during popular programmes, aiming to reach a broad audience. However, the limitations of traditional TV in providing detailed demographic insights pose challenges for advertisers in tailoring campaigns effectively, leads to resource inefficiencies and missed opportunities for precise audience engagement [5] . For instance, the declining interest of young adults in traditional TV poses targeting challenges for advertisers lacking specific demographic data [26] .  \nIntegrating first-party data sourced from Freeview television stands as a promising solution for advertisers grappling with limitations in traditional TV demographics. This data, acquired directly through viewers’ connected TVs using the standard Hybrid Broadcast Broadband TV, offers valuable insights into viewers’ habits and preferences, collected through their interactions with Freeview’s interactive services.  \n2 Carreno et al.  \nThis paper aims to utilise Machine Learning (ML) techniques to analyse this first-party data to construct accurate and comprehensive viewer profiles, bridging the gap in precise audience understanding that traditional TV lacks.  \n1.1 Paper’s contributions  \nThe contributions of this paper are:  \n- A Machine Learning-ready dataset with approximately 20,000 categorised devices from television first-party data. This dataset also includes a detailed household taxonomy and is made publicly available for further research.  \n- A detailed pipeline describing the entire process from initial data collection to the final classification of devices. This complete overview allows for an understanding of the systematic approach employed throughout the research.  \n- The baseline results achieved with state-of-the-art machine learning models to quantitatively demonstrate the feasibility of our approach.  \nThe rest of the paper is organised as follows. Section 2 elaborates the analytical pipeline to collect our dataset and extract insights from it. Section 4 details various ML algorithms and the baseline results on such dataset. Section 5 explains the related work. Finally, Section 6 summarises key insights and implications for future research.  \n2 Pipeline  \nThis section introduces a structured analytical pipeline involving 6 specific","cbCaimkL8LZfQejh","https://ap.wps.com/l/cbCaimkL8LZfQejh","pdf",1659888,1,21,"English","en",105,"# Introduction\n## Motivation and challenges in linear TV advertising\n## First-party data with HbbTV\n# Contributions\n## Machine learning-ready dataset\n## Analytical pipeline and baseline results\n# Pipeline\n## Data collection with HbbTV\n## Determining viewer time slots","[{\"question\":\"Why does linear television advertising face difficulties today?\",\"answer\":\"It suffers from fragmented viewership, limited demographic visibility, and higher costs, which makes targeting less precise and can waste resources.\"},{\"question\":\"What type of data is used to build viewer insights in this approach?\",\"answer\":\"The method uses television first-party data collected from connected Freeview TVs via HbbTV interactive services, including device and interaction context.\"},{\"question\":\"How accurate is the household demographic classification?\",\"answer\":\"State-of-the-art machine learning classification models achieve an average 88.7% accuracy in correctly identifying household demographics.\"}]","Unlocking Viewer Insights in Linear Television - A Machine Learning Approach | PDF",1785730379,53,{"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},"unlocking-viewer-insights-in-linear-television-a-machine-learning-approach","",{"@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/unlocking-viewer-insights-in-linear-television-a-machine-learning-approach/120500/",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","2026-08-04","2026-08-03",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 does linear television advertising face difficulties today?","Question",{"text":76,"@type":77},"It suffers from fragmented viewership, limited demographic visibility, and higher costs, which makes targeting less precise and can waste resources.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What type of data is used to build viewer insights in this approach?",{"text":81,"@type":77},"The method uses television first-party data collected from connected Freeview TVs via HbbTV interactive services, including device and interaction context.",{"name":83,"@type":74,"acceptedAnswer":84},"How accurate is the household demographic classification?",{"text":85,"@type":77},"State-of-the-art machine learning classification models achieve an average 88.7% accuracy in correctly identifying household demographics.","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"]