DC Field | Value | Language |
---|---|---|
dc.contributor.author | 이성제 | - |
dc.date.accessioned | 2022-03-29T03:23:39Z | - |
dc.date.available | 2022-03-29T03:23:39Z | - |
dc.date.issued | 2020 | - |
dc.identifier.other | OAK-2015-08859 | - |
dc.identifier.uri | http://postech.dcollection.net/common/orgView/200000336060 | ko_KR |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/111664 | - |
dc.description | Master | - |
dc.description.abstract | Session-based recommender systems aim to predict a user's next item using the previous behavior in a session. Sessions can vary in length, but previous studies have not focused on long sessions. This paper proposes a novel session-based recommender model that applies an attention method to a dilated recurrent neural network so that the model can effectively learn even if the session length is long. The dilated recurrent neural network in our model makes it possible to learn long-term dependency more effectively compared to the existing methods. The proposed method made the learning effective according to a scale of a session, and the experimental results showed that the model had the better performance than the existing models. | - |
dc.language | eng | - |
dc.publisher | 포항공과대학교 | - |
dc.title | Session-based Recommender System with Dilated Recurrent Neural Networks | - |
dc.type | Thesis | - |
dc.contributor.college | 일반대학원 컴퓨터공학과 | - |
dc.date.degree | 2020- 8 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.