HOME > Detail View

Detail View

Recommender system using sequential and global preference via attention mechanism and topic modeling

Recommender system using sequential and global preference via attention mechanism and topic modeling

Material type
학위논문
Personal Author
강경필, 康景弼
Title Statement
Recommender system using sequential and global preference via attention mechanism and topic modeling / Kyeong Pil Kang
Publication, Distribution, etc
Seoul :   Graduate School, Korea University,   2020  
Physical Medium
vi, 60장 : 삽화, 도표 ; 26 cm
기타형태 저록
Recommender System Using Sequential and Global Preference via Attention Mechanism and Topic Modeling   (DCOLL211009)000000127341  
학위논문주기
학위논문(박사)-- 고려대학교 대학원: 컴퓨터·전파통신공학과, 2020. 2
학과코드
0510   6YD36   377  
General Note
지도교수: 주재걸  
Bibliography, Etc. Note
참고문헌: 장 50-60
이용가능한 다른형태자료
PDF 파일로도 이용가능;   Requires PDF file reader(application/pdf)  
비통제주제어
Deep Learning, Topic Modeling, Recommender System,,
000 00000nam c2200205 c 4500
001 000046026272
005 20201116091647
007 ta
008 191231s2020 ulkad bmAC 000c eng
040 ▼a 211009 ▼c 211009 ▼d 211009
085 0 ▼a 0510 ▼2 KDCP
090 ▼a 0510 ▼b 6YD36 ▼c 377
100 1 ▼a 강경필, ▼g 康景弼
245 1 0 ▼a Recommender system using sequential and global preference via attention mechanism and topic modeling / ▼d Kyeong Pil Kang
260 ▼a Seoul : ▼b Graduate School, Korea University, ▼c 2020
300 ▼a vi, 60장 : ▼b 삽화, 도표 ; ▼c 26 cm
500 ▼a 지도교수: 주재걸
502 1 ▼a 학위논문(박사)-- ▼b 고려대학교 대학원: ▼c 컴퓨터·전파통신공학과, ▼d 2020. 2
504 ▼a 참고문헌: 장 50-60
530 ▼a PDF 파일로도 이용가능; ▼c Requires PDF file reader(application/pdf)
653 ▼a Deep Learning ▼a Topic Modeling ▼a Recommender System
776 0 ▼t Recommender System Using Sequential and Global Preference via Attention Mechanism and Topic Modeling ▼w (DCOLL211009)000000127341
900 1 0 ▼a Kang, Kyeong Pil, ▼e
900 1 0 ▼a 주재걸, ▼g 朱辛傑, ▼e 지도교수
900 1 0 ▼a Choo, Jae-gul, ▼e 지도교수
945 ▼a KLPA

Electronic Information

No. Title Service
1
Recommender system using sequential and global preference via attention mechanism and topic modeling (47회 열람)
View PDF Abstract Table of Contents

Holdings Information

No. Location Call Number Accession No. Availability Due Date Make a Reservation Service
No. 1 Location Science & Engineering Library/Stacks(Thesis)/ Call Number 0510 6YD36 377 Accession No. 123063739 Availability Available Due Date Make a Reservation Service B M
No. 2 Location Science & Engineering Library/Stacks(Thesis)/ Call Number 0510 6YD36 377 Accession No. 123063740 Availability Available Due Date Make a Reservation Service B M

Contents information

Abstract

As the size of the market in online social media and e-commerce has grown dramatically, the number of online commercial contents has increased at an exponential rate, and consequently, the recommender systems have been used to assist customers to explore their preferred products among lots of products in an efficient way. Furthermore, deep neural networks and an increase of training datasets have improved the accuracy of sequential recommendation approaches which take into account the sequential patterns of user logs, e.g., purchase histories of a user. 

However, incorporating only the individual's recent logs may not be sufficient in properly reflecting global preferences and trends across all users and items. In response, we propose a novel self-attentive sequential recommender system with topic modeling-based category embedding as a novel approach to exploit global information in the process of sequential recommendation. 

The self-attention module in our model effectively leverages the sequential patterns from the user's recent history. In addition, our novel category embedding approach, combined with the categorical probabilities calculated by topic modeling, effectively captures global information that the user generally prefers. We further propose a categorical preference gate to efficiently combine the sequential information and the global information for the given user's previous purchased logs or the user's general preference. Furthermore, as the diverse recommendation has become important for real-world applications, our model also incorporates a linearly-transformed noise vector obtained by random sampling so as to make users explore various items as well as to prevent loss in user information and the overfitting problem.

Experimental studies using public datasets, MovieLens and Gowalla, show that our model outperforms other state-of-the-art sequential recommendation models, and the results of the ablation test show that each module in our model plays important rule for improving the model performance. Additional qualitative experiments show that the proposed category embedding, combined with the proposed categorical preference gate, effectively provides global preference information. In addition, the linearly-transformed random noise vector added to the user embedding vector for personal recommendation effectively provides the diverse recommendation without loss in the user's general preference information, and the model further provides options to users for directly controlling the trade-off between their own preferences and exploring various items.

Table of Contents

1. Introduction 1
2. Related Work 5
 2.1 Recommender System 5
 2.2 Deep Learning and Attention Mechanism 7
 2.3 Topic Modeling 8
3. Proposed Methods 11
 3.1 Summary 11
 3.2 Item Encoding via Self-Attention Module 13
 3.3 Category Representation 16
 3.4 User Embedding 16
 3.5 Categorical Preference Gate 17
 3.6 Prediction Layer 18
 3.7 Optimization of Our Model 18
4. Experiments 20
 4.1 Datasets 20
 4.2 Experimental Setup 21
 4.3 Evaluation Metrics 22
 4.4 Baselines 23
5. Experimental Results 25
 5.1 Model Performance 25
 5.2 Ablation Test 30
 5.3 Visualization of Item Embedding 33
 5.4 Analysis of Category Embedding 33
 5.5 Analysis of Categorical Preference Gate 36
 5.6 Visualization of the Self-Attention Score 41
 5.7 Diverse Recommendation 42
  5.7.1 Impact of the degree of diversity 42
  5.7.2 Qualitative analysis of diverse recommendation 45
6. Conclusion 48
Bibliography 50