[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-582017-en":3,"detail-sidebar-cat-0-en-105":31,"doc-seo-582017-105":81},{"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,"preview_page_count":4},582017,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","Student-generated texts as features for predicting learning from video lectures - An initial evaluation","Digital trails on e-learning platforms have drawn major research interest, especially where students produce multiple kinds of texts. Although learners generate large volumes of writing, linguistic features derived from those texts have not been systematically studied for prediction. The paper proposes a framework using student-written summaries after short video lectures to predict learning performance. Forty-two undergraduates produced summaries, features were extracted, and multiple machine-learning classifiers were evaluated, showing promising results.","Student-generated texts as features for predicting learning from video lectures: An initial evaluation  \nIlias Karasavvidis, Charalampos Papadimas, Vasiliki Ragazou  \n[ikaras@uth.gr](ikaras@uth.gr), [papadimas@uth.gr](papadimas@uth.gr); [ragazou@uth.gr](ragazou@uth.gr)[ ](ragazou@uth.gr)Department of Early Childhood Education, University of Thessaly, Greece  \nAbstract  \nThe digital trails that students leave behind on e-learning environments have attracted considerable attention in the past decade. Typically, some of these traces involve the production of different kinds of texts. While students routinely produce a bulk of texts in online learning settings, the potential of such linguistic features has not been systematically explored. This paper introduces a novel approach that involves using student-generated texts for predicting performance after viewing short video lectures. Forty-two undergraduates viewed six video lectures and were asked to write short summaries for each one. Five combinations of features that were extracted from these summaries were used to train eight machine learning classifiers. The findings indicated that the raw text feature set achieved higher average classification accuracy in two video lectures, while the combined feature set whose dimensionality had been reduced resulted in higher classification accuracy in two other video lectures. The findings also indicated that the Gradient Boost, AdaBoost and Random Forest classifiers achieved high average performance in half of the video lectures. The study findings suggest that student-produced texts are a very promising source of features for predicting student performance when learning from short video lectures.  \nKeywords: machine learning, raw text features, engineered text features, video lectures, video learning analytics  \nIntroduction  \nNowadays, e-learning is ubiquitous, its adoption ranging from corporations to educational institutions. In the context of higher education both hybrid and fully online courses have become the norm. When studying online, students engage in various course activities such as lecture viewing, online readings, synchronous and asynchronous online discussions, quizzes, assessments, and assignment submissions (Ferguson, 2012; Hernández-García & Conde-González, 2016; Romero & Ventura, 2010; Tomasevic, Gvozdenovic & Vranes, 2020) . All student activity in such E-learning systems can be logged. Hence, students' interactions with the course materials are reflected in their digital footprints (Ifenthaler & Widanapathirana, 2014; Papamitsiou & Economides, 2014; Romero & Ventura, 2010; Schumacher & Ifenthaler, 2018) . Through their participation in e-learning environments, students generate huge amounts of digital data. The digital traces that the students leave have attracted the interest of educators, researchers, administrators, and other stakeholders.  \nOver the past decade, Educational Data Mining (EDM) and Learning Analytics (LA) emerged as research fields that aim to capitalize on such data (Ferguson, 2012) . According to Romero and Ventura (2020), LA can be defined as the measurement, collection, analysis, and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs. Video Learning Analytics (VLA) is a sub-set of LA that focuses on students’ learning behaviours when learning from video lectures (Giannakos, Chorianopoulos & Chrisochoides, 2014; Hasan et al., 2020) .  \nThe value of the data that is collected through online learning systems is widely recognized. The first wave of studies in the field focused on the identification of variables that could be used to predict student performance, increase student retention, and identify at risk students.  \nMore specifically, LA studies have employed various measures such as (a) demographic data: age, date of birth, citizenship, prior academic history, expertise (Stöhr, Stathakarou, Mueller","cbCailtPzhZ6oQlr","https://ap.wps.com/l/cbCailtPzhZ6oQlr","pdf",1736450,2,1,25,"English","en",105,"# Abstract\n# Introduction\n## E-learning data and digital footprints\n## Educational Data Mining and Learning Analytics\n## Predictive features in Learning Analytics\n## Student-produced texts and research gap","[{\"question\":\"What data source does the paper use to predict learning from video lectures?\",\"answer\":\"The study uses student-generated texts, specifically short summaries written after viewing six short video lectures.\"},{\"question\":\"How were machine learning models trained and evaluated?\",\"answer\":\"Five combinations of extracted text features were used to train eight machine learning classifiers, and their classification accuracy was compared across the lectures.\"},{\"question\":\"What overall conclusion does the study reach about student-produced texts?\",\"answer\":\"Student-produced texts are presented as a promising source of features for predicting student performance when learning from short video lectures.\"}]","Student-generated texts as features for predicting learning from video lectures - An initial evaluation | PDF",1792042460,63,{"code":4,"msg":5,"data":32},[33,38,42,47,52,57,62,65,70,73,77],{"id":21,"doc_module":4,"doc_module_name":34,"category_name":35,"show_sort_weight":36,"slug":37},"Document","Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":34,"category_name":39,"show_sort_weight":40,"slug":41},"Literature",80,"literature",{"id":43,"doc_module":4,"doc_module_name":34,"category_name":44,"show_sort_weight":45,"slug":46},4,"Exam",70,"exam",{"id":48,"doc_module":4,"doc_module_name":34,"category_name":49,"show_sort_weight":50,"slug":51},5,"Comic",60,"comic",{"id":53,"doc_module":4,"doc_module_name":34,"category_name":54,"show_sort_weight":55,"slug":56},6,"Technology",50,"technology",{"id":58,"doc_module":4,"doc_module_name":34,"category_name":59,"show_sort_weight":60,"slug":61},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":34,"category_name":12,"show_sort_weight":63,"slug":64},30,"research-report",{"id":66,"doc_module":4,"doc_module_name":34,"category_name":67,"show_sort_weight":68,"slug":69},9,"Religion & Spirituality",20,"religion-spirituality",{"id":68,"doc_module":4,"doc_module_name":34,"category_name":71,"show_sort_weight":68,"slug":72},"World Cup","world-cup",{"id":74,"doc_module":4,"doc_module_name":34,"category_name":75,"show_sort_weight":74,"slug":76},10,"Lifestyle","lifestyle",{"id":78,"doc_module":4,"doc_module_name":34,"category_name":79,"show_sort_weight":48,"slug":80},19,"General","general",{"code":4,"msg":82,"data":83},"ok",{"site_id":25,"language":24,"slug":84,"title":13,"keywords":85,"description":14,"schema_data":86,"social_meta":140,"head_meta":142,"extra_data":144,"updated_unix":145},"student-generated-texts-as-features-for-predicting-learning-from-video-lectures-an-initial-evaluation","",{"@graph":87,"@context":139},[88,102,122],{"@type":89,"itemListElement":90},"BreadcrumbList",[91,95,97,100],{"item":92,"name":93,"@type":94,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":96,"name":34,"@type":94,"position":20},"https://docshare.wps.com/document/",{"item":98,"name":12,"@type":94,"position":99},"https://docshare.wps.com/document/research-report/",3,{"item":101,"name":13,"@type":94,"position":43},"https://docshare.wps.com/document/student-generated-texts-as-features-for-predicting-learning-from-video-lectures-an-initial-evaluation/582017/",{"url":101,"name":13,"@type":103,"image":104,"author":109,"headline":13,"publisher":111,"fileFormat":114,"inLanguage":24,"description":14,"dateModified":115,"datePublished":116,"encodingFormat":114,"isAccessibleForFree":117,"interactionStatistic":118},"DigitalDocument",{"url":105,"@type":106,"width":107,"height":108},"https://docshare.wps.com/thumbnails/student-generated-texts-as-features-for-predicting-learning-from-video-lectures-an-initial-evaluation/582017.png","ImageObject",300,407,{"name":9,"@type":110},"Person",{"url":92,"name":112,"@type":113},"DocShare","Organization","application/pdf","2026-11-09","2026-10-15",true,{"@type":119,"interactionType":120,"userInteractionCount":20},"InteractionCounter",{"@type":121},"ViewAction",{"@type":123,"mainEntity":124},"FAQPage",[125,131,135],{"name":126,"@type":127,"acceptedAnswer":128},"What data source does the paper use to predict learning from video lectures?","Question",{"text":129,"@type":130},"The study uses student-generated texts, specifically short summaries written after viewing six short video lectures.","Answer",{"name":132,"@type":127,"acceptedAnswer":133},"How were machine learning models trained and evaluated?",{"text":134,"@type":130},"Five combinations of extracted text features were used to train eight machine learning classifiers, and their classification accuracy was compared across the lectures.",{"name":136,"@type":127,"acceptedAnswer":137},"What overall conclusion does the study reach about student-produced texts?",{"text":138,"@type":130},"Student-produced texts are presented as a promising source of features for predicting student performance when learning from short video lectures.","https://schema.org",{"og:url":101,"og:type":141,"og:title":13,"og:site_name":112,"og:description":14},"article",{"robots":143,"canonical":101},"index,follow",{"doc_id":7,"site_id":25},1792209182]