HOME > 상세정보

상세정보

Improving point-of-interest recommendation with Text Mining from User Generated Content

Improving point-of-interest recommendation with Text Mining from User Generated Content

자료유형
학위논문
개인저자
Chang, Buru
서명 / 저자사항
Improving point-of-interest recommendation with Text Mining from User Generated Content / Buru Chang 저
발행사항
Seoul :   Graduate School, Korea University,   2020  
형태사항
vi, 72장 : 천연색삽화, 도표 ; 26 cm
기타형태 저록
Improving Point-of-Interest Recommendation with Text Mining from User Generated Content   (DCOLL211009)000000232145  
학위논문주기
학위논문(박사)-- 고려대학교 대학원: 컴퓨터·전파통신공학과, 2020. 8
학과코드
0510   6YD36   384  
일반주기
지도교수: 강재우  
서지주기
참고문헌: 장 66-72
이용가능한 다른형태자료
PDF 파일로도 이용가능;   Requires PDF file reader(application/pdf)  
비통제주제어
Text Mining , Point-of-Interest Recommendation,,
000 00000nam c2200205 c 4500
001 000046048414
005 20200921092258
007 ta
008 200630s2020 ulkad bmAC 000c eng
040 ▼a 211009 ▼c 211009 ▼d 211009
085 0 ▼a 0510 ▼2 KDCP
090 ▼a 0510 ▼b 6YD36 ▼c 384
100 1 ▼a Chang, Buru
245 1 0 ▼a Improving point-of-interest recommendation with Text Mining from User Generated Content / ▼d Buru Chang 저
260 ▼a Seoul : ▼b Graduate School, Korea University, ▼c 2020
300 ▼a vi, 72장 : ▼b 천연색삽화, 도표 ; ▼c 26 cm
500 ▼a 지도교수: 강재우
502 1 ▼a 학위논문(박사)-- ▼b 고려대학교 대학원: ▼c 컴퓨터·전파통신공학과, ▼d 2020. 8
504 ▼a 참고문헌: 장 66-72
530 ▼a PDF 파일로도 이용가능; ▼c Requires PDF file reader(application/pdf)
653 ▼a Text Mining ▼a Point-of-Interest Recommendation
776 0 ▼t Improving Point-of-Interest Recommendation with Text Mining from User Generated Content ▼w (DCOLL211009)000000232145
900 1 0 ▼a 강재우, ▼g 姜在雨, ▼e 지도교수
900 1 0 ▼a 장부루, ▼e
945 ▼a KLPA

전자정보

No. 원문명 서비스
1
Improving point-of-interest recommendation with Text Mining from User Generated Content (26회 열람)
PDF 초록 목차

소장정보

No. 소장처 청구기호 등록번호 도서상태 반납예정일 예약 서비스
No. 1 소장처 과학도서관/학위논문서고/ 청구기호 0510 6YD36 384 등록번호 123064854 도서상태 대출가능 반납예정일 예약 서비스 B M
No. 2 소장처 과학도서관/학위논문서고/ 청구기호 0510 6YD36 384 등록번호 123064855 도서상태 대출가능 반납예정일 예약 서비스 B M

컨텐츠정보

초록

Over the last decade, mobile-based social media platforms such as Twitter, Facebook, Instagram, and Foursquare have grown in popularity. Users on these platforms generate a large amount of check-in histories which include temporal and spatial information. Such information is particularly useful for understanding user behavior patterns and preferences for points-of-interest (POIs) that are specific locations that someone finds interesting. POI recommendation that recommends POIs for users to visit next based on the check-in histories plays an important role in mobile-based social media platforms. Users on these platforms also generate textual content on their experiences at POIs. User-generated textual content can be used for understanding POI characteristics and user interests. However, most of the existing POI recommendation methods highly focus on geographical influences among POIs, but they do not utilize user-generated textual content.

In this dissertation, we introduce our text mining studies for improving the performance of POI recommendation methods from user-generated content. We first propose a POI imputation model that utilizes user-generated content to automatically tag POI information in social media posts. Furthermore, we propose a content-aware POI embedding method that captures not only geographical influences among POIs but also characteristics of POIs from user-generated textual content. The proposed POI embedding method improves the performance of successive POI recommendation as latent representations of POIs of the successive POI recommendation models. Last, we propose a content-aware successive POI recommendation model that uses user-generated textual content to capture user interests. Through the experimental evaluation on real-world datasets, we show that our proposed models in each study achieve state-of-the-art performance in their tasks, respectively.

목차

Abstract

Contents

1 Introduction 1

2 POI Imputation 5
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1 Point-of-Interest imputation . . . . . . . . . . . . . . . . . . . . . . 9
2.2.2 Image Geolocation Estimation . . . . . . . . . . . . . . . . . . . . 9
2.2.3 Point-of-Interest Recommendation . . . . . . . . . . . . . . . . . . 10
2.3 Our Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.1 Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.2 DeepPIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.4 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4.1 Performance Comparison with Baseline models . . . . . . . . . . . 19
2.4.2 Evaluation of Our Model’s Performance on our Large-scale Dataset 22
2.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3 User-generated Content-aware POI Embedding 27
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2.1 Successive POI Recommendation . . . . . . . . . . . . . . . . . . . 30
3.2.2 POI Embedding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3 Data Description and Analysis . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3.1 Data Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3.2 Empirical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.4 Our Appraoch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4.1 Check-in Context Layer . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4.2 Text Content Layer . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.5 Experimental Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.5.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.5.2 Evaluation Result . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

4 User-generated Content-aware POI Recommendation 44
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.2.1 Successive POI Recommendation . . . . . . . . . . . . . . . . . . . 47
4.2.2 Content-aware POI Recommendation . . . . . . . . . . . . . . . . 47
4.3 Our Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.3.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.3.2 Content Encoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.3.3 CAPRE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.4 Experimental Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.4.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.4.2 Evaluation Result . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

5 Conclusion 64
Bibliography 66