Browse

EFFICIENT TRANSFER LEARNING SCHEMES FOR PERSONALIZED LANGUAGE MODELING USING RECURRENT NEURAL NETWORK
순환신경망을 이용한 개인화 언어모델 학습 방법에 관한 연구

DC Field Value Language
dc.contributor.advisor정교민-
dc.contributor.author윤승현-
dc.date.accessioned2017-07-14T02:43:41Z-
dc.date.available2017-07-14T02:43:41Z-
dc.date.issued2017-02-
dc.identifier.other000000140914-
dc.identifier.urihttps://hdl.handle.net/10371/122836-
dc.description학위논문 (석사)-- 서울대학교 대학원 : 전기·정보공학부, 2017. 2. 정교민.-
dc.description.abstractIn this paper, we propose an efficient transfer leaning methods for training a personalized language model using a recurrent neural network with long short-term memory architecture. With our proposed fast transfer learning schemes, a general language model is updated to a personalized language model with a small amount of user data and a limited computing resource. These methods can be applied especially useful to a mobile device environment while the data is prevented from transferring out of the device for privacy purposes. Through experiments on dialogue data in a drama, it is verified that our transfer learning methods have successfully generated the personalized language model.-
dc.description.tableofcontents1 INTRODUCTION 1
2 RELATED WORK 3
3 METHODOLOGY 5
3.1 Language Model 5
3.1.1 Sentence Completion Language Model 6
3.1.2 Message-Reply Prediction Language Model 6
3.2 Proposed Methods 7
3.2.1 Fast Transfer Learning Schemes 7
3.3 Measures 8
3.3.1 Cross Entropy Metric 8
4 EXPERIMENTS 11
4.1 Datasets 11
4.2 Test Results 12
4.2.1 Literary-Style to Spoken-Style Sentence Completion 12
4.2.2 General-Style to Personal-Style Message-Reply Prediction 13
5 CONCLUSION 20
Abstract (In Korean) 24
-
dc.formatapplication/pdf-
dc.format.extent2076815 bytes-
dc.format.mediumapplication/pdf-
dc.language.isoen-
dc.publisher서울대학교 대학원-
dc.subject자연어처리-
dc.subject딥러닝-
dc.subject순환신경망모델-
dc.subject개인정보보호-
dc.subject언어모델-
dc.subject.ddc621-
dc.titleEFFICIENT TRANSFER LEARNING SCHEMES FOR PERSONALIZED LANGUAGE MODELING USING RECURRENT NEURAL NETWORK-
dc.title.alternative순환신경망을 이용한 개인화 언어모델 학습 방법에 관한 연구-
dc.typeThesis-
dc.description.degreeMaster-
dc.citation.pages24-
dc.contributor.affiliation공과대학 전기·정보공학부-
dc.date.awarded2017-02-
Appears in Collections:
College of Engineering/Engineering Practice School (공과대학/대학원)Dept. of Electrical and Computer Engineering (전기·정보공학부)Theses (Master's Degree_전기·정보공학부)
Files in This Item:
  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse