생몰정보
소속
직위
직업
활동분야
주기
서지
국회도서관 서비스 이용에 대한 안내를 해드립니다.
검색결과 (전체 1건)
원문 있는 자료 (1) 열기
원문 아이콘이 없는 경우 국회도서관 방문 시 책자로 이용 가능
목차보기더보기
Title Page
ABSTRACT
Abstract (in Korean)
Contents
Chapter 1. Introduction 13
1.1. Problem Statement 14
1.1.1. Short length of messages 14
1.1.2. Absence of category information 15
1.2. Proposed Approach 16
1.3. Outline 19
Chapter 2. Background and Related Works 21
2.1. Definitions and Background Theories 21
2.2. Literature Review 29
2.2.1. Personalization in Social Media 29
2.2.2. Short Text Classification 33
2.2.3. Leveraging Wikipedia for Text Classification 34
2.2.4. Deep Learning for Natural Language Processing 37
Chapter 3. Unsupervised Methods 41
3.1. The architecture of our proposed system 41
3.2. Preliminary 43
3.3. Term-based feature generator 43
3.4. Wikipedia-based feature generators 46
3.4.1. Structure of Wikipedia 46
3.4.2. Wiki-CF-ICF feature generator 47
3.4.3. Wiki-Cluster feature generator 52
3.4.4. Wiki-AF-IAF feature generator 56
3.4.5. Wiki-Mix that integrates Wikipedia-based features 59
3.5. User profiling and category recommendation using Wikipedia-based feature generators 59
Chapter 4. Supervised Methods 62
4.1. The architecture of the proposed approach 62
4.2. Convolutional Neural Networks 63
4.3. Learning the Model Parameters 66
Chapter 5. Evaluations 68
5.1. Data sets 68
5.2. Accuracy of users' estimated interests in an unsupervised setting 69
5.2.1. Evaluation framework 69
5.2.2. Performance measure 70
5.2.3. Parameter estimation 72
5.2.4. Results 72
5.3. Accuracy of users' estimated interests in a supervised setting 79
5.3.1. Evaluation framework 79
5.3.2. Performance measure 79
5.3.3. Results 81
Chapter 6. Conclusion and Future Work 84
References 87
Curriculum Vitae 100
Table 2.1: A particla list of social media site 22
Table 2.2: Top 10 sites ordered by internet traffic 22
Table 3.1: Categories and subcategories of news media 44
Table 3.2: Notations and descriptions of notations 44
Table 5.1: Environmental setting for parameter estimation 73
Table 5.2: The results of statistical significant tests 74
Table 5.3: Messages of Facebook users and results of categorization using TF-ICF and... 78
Table 5.4: Message-level and user-level accuracies 82
Figure 1.1: The overall flow diagram of our proposed approaches for modeling user interest 16
Figure 2.1: Layers in a Convolutional Neural Network. A CNN is a succession of Convo-... 27
Figure 2.2: The Convolution Process. This figure shows the process of convolving a 3x3... 28
Figure 2.3: The Subsampling Step in CNN. This figure shows the process of subsampling... 28
Figure 2.4: Flow diagram of a tweet categorization method using pins and tweets 30
Figure 2.5: Entity based topic profiles using Wikipedia 35
Figure 2.6: Architecture of hierarchical interest identification using Wikipedia 36
Figure 2.7: Two vector representation of words models. CBOW predicts the current word... 38
Figure 3.1: A framework of our system 42
Figure 3.2: Structure of Wikipedia 46
Figure 3.3: A Wiki-CF-ICF feature generator using a Wiki-category generation module... 47
Figure 3.4: Wiki-Cluster feature generator using clustering method 52
Figure 3.5: Wiki-AF-IAF feature generator using semantic interpreter 56
Figure 4.1: The architecture of the CNN used in our approach. 63
Figure 5.1: Average accuracies of all users with threshold values of thresholdcategoryscore(이미지참조) 77
Figure 5.2: Average top-k accuracies of all users (thresholdcategoryscore=optimal scores...(이미지참조) 77
원문구축 및 2018년 이후 자료는 524호에서 직접 열람하십시요.
도서위치안내: / 서가번호:
우편복사 목록담기를 완료하였습니다.
* 표시는 필수사항 입니다.
* 주의: 국회도서관 이용자 모두에게 공유서재로 서비스 됩니다.
저장 되었습니다.
로그인을 하시려면 아이디와 비밀번호를 입력해주세요. 모바일 간편 열람증으로 입실한 경우 회원가입을 해야합니다.
공용 PC이므로 한번 더 로그인 해 주시기 바랍니다.
아이디 또는 비밀번호를 확인해주세요