[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128030-en":3,"doc-seo-128030-105":30,"detail-sidebar-cat-0-en-105":91},{"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":11,"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},128030,962084931830,"Theodore","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","BERT-RS - A Neural Personalized Recommender System with BERT","Accurate user preferences and item representations are fundamental to personalized recommender systems, yet collaborative filtering often underuses sentiment and preference signals contained in review text. BERT-RS introduces a neural personalized recommendation model that extracts semantic representations from textual reviews using BERT, then derives user and item latent representations via three deep architectures. Personalized recommendations are produced through score prediction on the learned representations, delivering strong experimental results on the Amazon dataset compared with baseline methods.","September 12, 2022 15:27 WSPC Proceedings-9in x 6in ws-procs9x6 page 1  \n1  \nBERT-RS: A Neural Personalized Recommender System with  \nBERT  \nKezhi Lu, Qian Zhang, Guangquan Zhang and Jie Lu  \nAustralian Arti􀀌cial Intelligence Institute, Faulty of Engineering and Information Technology,  \nUniversity of Technology Sydney,  \nSydney, NSW, 2007, Australia  \nE-mail: [lukezhi@bjtu.edu.cn](lukezhi@bjtu.edu.cn); fqian.zhang-1, guangquan.zhang, [jie.lu](jie.lug@uts.edu.au)[g](jie.lug@uts.edu.au)[@uts.edu.au](jie.lug@uts.edu.au)  \nAccurate user preferences and item representations are essential factors for personalized recommender systems. Explicit feedback behaviors, such as ratings and free-text comments, are rich in personalized preference knowledge and emotional evaluation information. It is a direct and e􀀋ective way to obtain individualized preference and item latent representations from these sources. In this paper, we propose a novel neural model named BERT-RS for personalized recommender systems, which extracts knowledge from textual reviews and user-item interactions. First, we preliminary extract the semantic representation for users and items from the textual comments based on BERT. Next, these semantic embeddings are used for user and item latent representations through three di􀀋erent deep architectures. Finally, we carry out personalized recommendation tasks through the score prediction based on these representations. Compared with other algorithms, BERT-RS demonstrates outstanding  \nexperimental performance on the Amazon dataset.  \nKeywords: Collaborative Filtering; Recommender Systems; BERT; Personal  \nized Recommendation.  \n1. INTRODUCTION  \nIn the information age, the recommender system (RS) is crucial to solving the problem of information overload in Internet services. Providing personalized recommendations is the central goal for the RS. Meanwhile, the core in the personalized RS is accurately identifying the user's preferences and linking them with items or online services 1 . To this day, the collaborative 􀀌ltering (CF) method has been used to obtain and predict users'similar preferences through the past interaction information between users and items (e.g., click and rating), which has been successfully and widely used in RS 2 .  \nCF-based algorithms recommend relevant products according to users'  \nSeptember 12, 2022 15:27 WSPC Proceedings-9in x 6in ws-procs9x6 page 2  \n2  \nsimilar preferences 3{5 . The CF-based model generally has two key points:  \n(1) construct the similarity representation between users and items according to the past interaction records; (2) construct the model to train and recommend relevant items. For example, the neural collaborative 􀀌ltering (NCF) method uses nonlinear neural networks with di􀀋erent structures based on matrix factorization (MF) to construct the representation of users and items 3,6 . However, the NCF model can not deeply analyze the potential preference information and emotional factors according to textual comments. As a result, in this paper, we propose a novel model utilizing reviews knowledge and user-item interactions to construct user/item representations for personalized recommendations.  \nReview texts contain rich knowledge of personal preferences and sentiments, which can facilitate the accurate representation of users and items. In NLP, various BERT-based models show their state-of-the-art performance in multiple tasks such as Named Entity Recognition (NER) and Question Answering (QA) 7 . Inspired by this, we fuse the BERT with neural architectures to extract semantic embeddings from reviews content and construct the personalized user/item representations based on them for RS. In general, BERT-RS has three di􀀋erent parts. Firstly, we extract and  \n􀀌ne-tune the preliminary semantic embeddings from reviews based on the BERT. Secondly, we fuse semantic information to construct user and item personalized representations through three di􀀋erent neural architectures, i.e.","cbCainPs2Wv8VTG7","https://ap.wps.com/l/cbCainPs2Wv8VTG7","pdf",933516,2,1,"English","en",105,"# Introduction\n## Motivation for personalized recommendation\n## Collaborative filtering and limitations\n## Role of review text and BERT\n# Related Work\n## Collaborative filtering and matrix factorization\n## Neural collaborative filtering\n## BERT and pre-trained language representations","[{\"question\":\"What problem does BERT-RS address in personalized recommender systems?\",\"answer\":\"It targets the challenge of accurately identifying user preferences and connecting them to items while leveraging review text that contains both preference and sentiment information.\"},{\"question\":\"How does BERT-RS use BERT in the recommendation pipeline?\",\"answer\":\"It extracts semantic representations for users and items from textual comments by using BERT, then fine-tunes and uses these embeddings to form user/item latent representations.\"},{\"question\":\"What model components does BERT-RS employ for user and item representations?\",\"answer\":\"It fuses semantic information through three neural architectures—BERT, GMF, and MLP—to construct personalized representations used for score prediction.\"}]","BERT-RS - A Neural Personalized Recommender System with BERT | PDF",1785944148,20,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":28},"bert-rs-a-neural-personalized-recommender-system-with-bert","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/bert-rs-a-neural-personalized-recommender-system-with-bert/128030/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does BERT-RS address in personalized recommender systems?","Question",{"text":75,"@type":76},"It targets the challenge of accurately identifying user preferences and connecting them to items while leveraging review text that contains both preference and sentiment information.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does BERT-RS use BERT in the recommendation pipeline?",{"text":80,"@type":76},"It extracts semantic representations for users and items from textual comments by using BERT, then fine-tunes and uses these embeddings to form user/item latent representations.",{"name":82,"@type":73,"acceptedAnswer":83},"What model components does BERT-RS employ for user and item representations?",{"text":84,"@type":76},"It fuses semantic information through three neural architectures—BERT, GMF, and MLP—to construct personalized representations used for score prediction.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":29,"slug":126},9,"Religion & Spirituality","religion-spirituality",{"id":29,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":29,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]