[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-128856-105":59,"doc-detail-128856-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","collaborative-recurrent-neural-networks-for-dynamic-recommender-systems-workshop-and-conference-proceedings-60","Collaborative Recurrent Neural Networks for Dynamic Recommender Systems - Workshop and Conference Proceedings 60","","Modern technologies record sequences of online user activity at unprecedented scale, while most recommender systems still focus on rating-prediction and overlook temporal and contextual behavior patterns. This work leverages non-intrusive activity logs to model dynamic preferences by combining latent factor ideas from collaborative filtering with language-modeling concepts. A recurrent neural sequence model captures a user’s contextual state as a personalized hidden vector summarizing variable-length past steps and represents items with real-valued embeddings. Experiments on music recommendation and mobility prediction show consistent improvements over static and non-collaborative methods.",{"@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/collaborative-recurrent-neural-networks-for-dynamic-recommender-systems-workshop-and-conference-proceedings-60/128856/",{"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/collaborative-recurrent-neural-networks-for-dynamic-recommender-systems-workshop-and-conference-proceedings-60/128856.png","ImageObject",300,407,{"name":92,"@type":93},"Aria","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-09-18","2026-08-06",true,{"@type":102,"interactionType":103,"userInteractionCount":52},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why is rating-prediction insufficient for modeling users in this work?","Question",{"text":112,"@type":113},"Explicit rating prediction is criticized for relying on one-time ratings and ignoring the immediate context and temporal dynamics of user preferences.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What data source does the proposed approach use for recommendations?",{"text":117,"@type":113},"The approach uses non-intrusive implicit activity logs, i.e., sequences of user interactions, which better support dynamic user modeling.",{"name":119,"@type":110,"acceptedAnswer":120},"How does the model represent users and items?",{"text":121,"@type":113},"It captures a user’s contextual state as a personalized hidden vector summarizing a variable number of past time steps, while items are represented through real-valued embeddings.","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},128856,1786003941,{"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":52,"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":36},2336474459895,"https://ap-avatar.wpscdn.com/avatar/22000baeef7a5ed0655?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786071322749376916","View metadata, citation and similar [papers at ](papers at core.ac.uk)[core.ac.uk](papers at core.ac.uk) brought to you by CORE  \nprovided by Infoscience- École polytechnique fédérale de Lausanne  \nJMLR: Workshop and Conference Proceedings 60:1{16, 2016 ACML 2016  \nCollaborative Recurrent Neural Networks for Dynamic Recommender Systems  \nYoung-Jun Ko [youngjun.ko@epfl.ch](youngjun.ko@epfl.ch)  \nLucas Maystre [lucas.maystre@epfl.ch](lucas.maystre@epfl.ch)  \nMatthias Grossglauser [matthias.grossglauser@epfl.ch](matthias.grossglauser@epfl.ch)  \n􀀓  \nEcole Polytechnique F􀀓ed􀀓erale de Lausanne, Switzerland  \nEditors: Robert J. Durrant and Kee-Eung Kim  \nAbstract  \nModern technologies enable us to record sequences of online user activity at an unprecedented scale. Although such activity logs are abundantly available, most approaches to recommender systems are based on the rating-prediction paradigm, ignoring temporal and contextual aspects of user behavior revealed by temporal, recurrent patterns. In contrast to explicit ratings, such activity logs can be collected in a non-intrusive way and can o􀀋er richer insights into the dynamics of user preferences, which could potentially lead more accurate user models.  \nIn this work we advocate studying this ubiquitous form of data and, by combining ideas from latent factor models for collaborative 􀀌ltering and language modeling, propose a novel, 􀀍exible and expressive collaborative sequence model based on recurrent neural networks. The model is designed to capture a user's contextual state as a personalized hidden vector by summarizing cues from a data-driven, thus variable, number of past time steps, and represents items by a real-valued embedding. We found that, by exploiting the inherent structure in the data, our formulation leads to an e􀀎cient and practical method. Furthermore, we demonstrate the versatility of our model by applying it to two di􀀋erent tasks: music recommendation and mobility prediction, and we show empirically that our model consistently outperforms static and non-collaborative methods.  \nKeywords: Recurrent Neural Network, Recommender System, Neural Language Model, Collaborative Filtering  \n1. Introduction  \nAs ever larger parts of the population routinely consume online an increasing amount of digital goods and services, and with the proliferation of a􀀋ordable storage and computing resources, to gain valuable insight, content-, product- and service-providing organizations face the challenge and the opportunity of tracking at a large scale the activity of their users. Modeling the underlying mechanisms that govern a users' choice for a particular item at a particular time is useful for, e.g., boosting user engagement or sales by assisting users in navigating overwhelmingly large product catalogs through recommendations, or by using pro􀀌ts from accurately targeted advertisement to keep services free of charge. Developing appropriate methodologies that use the available data e􀀋ectively is therefore of great practical interest.  \n􀀍c 2016 Y.-J. Ko, L. Maystre & M. Grossglauser.  \nKo Maystre Grossglauser  \nIn recent years, the recommendation problem has often been cast into an explicit-ratingprediction problem, possibly facilitated by the availability of appropriate datasets (e.g. , Bennett and Lanning, 2007; Miller et al. , 2003) . Incorporating a temporal aspect into the explicit-rating recommendation scenario is an active area of research (e.g. , Rendle, 2010; Koren, 2010; Koenigstein et al. , 2011; Chi and Kolda, 2012) . However, concerns are increasingly being raised about the suitability of explicit rating prediction as an e􀀋ective paradigm for user modeling. Featuring prominently among the criticism are concerns about the availability and reliability of explicit ratings, as well as the static nature of this paradigm (Yi et al. , 2014; Du et al. , 2015), in which the tastes and interests of users are assumed tobe captured by a one-time rating, thus neglecting the imme","cbCaijekOEoT74My","https://ap.wps.com/l/cbCaijekOEoT74My","pdf",471349,16,"English","# Introduction\n## Preliminaries","[{\"question\":\"Why is rating-prediction insufficient for modeling users in this work?\",\"answer\":\"Explicit rating prediction is criticized for relying on one-time ratings and ignoring the immediate context and temporal dynamics of user preferences.\"},{\"question\":\"What data source does the proposed approach use for recommendations?\",\"answer\":\"The approach uses non-intrusive implicit activity logs, i.e., sequences of user interactions, which better support dynamic user modeling.\"},{\"question\":\"How does the model represent users and items?\",\"answer\":\"It captures a user’s contextual state as a personalized hidden vector summarizing a variable number of past time steps, while items are represented through real-valued embeddings.\"}]","Collaborative Recurrent Neural Networks for Dynamic Recommender Systems - Workshop and Conference Proceedings 60 | PDF"]