Please use this identifier to cite or link to this item: https://dspace.ctu.edu.vn/jspui/handle/123456789/109468
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorNguyễn, Thái Nghe-
dc.contributor.authorNguyễn, Tú Trinh-
dc.date.accessioned2024-12-23T02:30:44Z-
dc.date.available2024-12-23T02:30:44Z-
dc.date.issued2024-
dc.identifier.otherB2015018-
dc.identifier.urihttps://dspace.ctu.edu.vn/jspui/handle/123456789/109468-
dc.description64 Trvi_VN
dc.description.abstractThe rapid growth of data requires systems capable of extracting relevant insights and delivering effective recommendations. Traditional recommendation approaches often face challenges in understanding the relationships and contextual dependencies within large datasets. This creates a need for more robust methods to address these limitations. BERT (Bidirectional Encoder Representations from Transformers) is a language model designed to capture contextual and semantic relationships in text. Its bidirectional processing enables a better understanding of data compared to conventional methods. Integrating BERT into recommendation systems can improve the accuracy and relevance of recommendations by leveraging its capacity for deep context understanding. This study examines the use of BERT in recommendation systems, focusing on its ability to address challenges in traditional approaches. The research evaluates the performance of BERT-based models and their impact on recommendation accuracy and system effectiveness. The findings aim to provide insights into the use of advanced language models for improving recommendation systems.vi_VN
dc.language.isoenvi_VN
dc.publisherTrường Đại Học Cần Thơvi_VN
dc.subjectCÔNG NGHỆ THÔNG TIN - CHẤT LƯỢNG CAOvi_VN
dc.titleA PRODUCT REMCOMMENDATION SYSTEM USING BERT4REC MODEL.vi_VN
dc.title.alternativeHỆ THỐNG GỢI Ý SẢN PHẨM SỬ DỤNG MÔ HÌNH BERT4RECvi_VN
dc.typeThesisvi_VN
Appears in Collections:Trường Công nghệ Thông tin & Truyền thông

Files in This Item:
File Description SizeFormat 
_file_
  Restricted Access
2.45 MBAdobe PDF
Your IP: 3.133.148.222


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.