An efficient location recommendation scheme based on clustering and data fusion

Wenjie Cai, Yufeng Wang, Ruheng Lv, Qun Jin

Research output: Contribution to journalArticle

Abstract

Recently, location-based social networks (LBSNs) have witnessed a significant development, and have attracted millions of mobile users to share their locations and location-related contents. Due to the huge data in LBSNs, it is imperative to design an efficient location recommendation scheme for abundant users. But most of existing works suffer from poor accuracy and efficiency. On one hand, the traditional user-based collaborative filtering (CF) methods only focus on user characteristics, which limit the recommendation accuracy. On the other hand, it is inefficient to traverse all LBSN data in each recommendation. To solve the above issues, this paper proposes a new location recommendation method LC–G–P, in which, Louvain method is adopted to cluster users into several communities; and meanwhile, multiple features including Geographical distance, location Popularity are incorporated into location recommendation. Experiments based on real LBSN dataset illustrate that the proposed recommendation scheme LC–G–P has better accuracy and efficiency, in comparison with the existed typical location recommendation schemes.

Original languageEnglish
Pages (from-to)289-299
Number of pages11
JournalComputers and Electrical Engineering
Volume77
DOIs
Publication statusPublished - 2019 Jul 1

Fingerprint

Data fusion
Collaborative filtering

Keywords

  • Collaborative filtering
  • Community detection
  • Data fusion
  • Location recommendation

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science(all)
  • Electrical and Electronic Engineering

Cite this

An efficient location recommendation scheme based on clustering and data fusion. / Cai, Wenjie; Wang, Yufeng; Lv, Ruheng; Jin, Qun.

In: Computers and Electrical Engineering, Vol. 77, 01.07.2019, p. 289-299.

Research output: Contribution to journalArticle

@article{7d6fcaca5cf9473db6d0a1a8106cf5a4,
title = "An efficient location recommendation scheme based on clustering and data fusion",
abstract = "Recently, location-based social networks (LBSNs) have witnessed a significant development, and have attracted millions of mobile users to share their locations and location-related contents. Due to the huge data in LBSNs, it is imperative to design an efficient location recommendation scheme for abundant users. But most of existing works suffer from poor accuracy and efficiency. On one hand, the traditional user-based collaborative filtering (CF) methods only focus on user characteristics, which limit the recommendation accuracy. On the other hand, it is inefficient to traverse all LBSN data in each recommendation. To solve the above issues, this paper proposes a new location recommendation method LC–G–P, in which, Louvain method is adopted to cluster users into several communities; and meanwhile, multiple features including Geographical distance, location Popularity are incorporated into location recommendation. Experiments based on real LBSN dataset illustrate that the proposed recommendation scheme LC–G–P has better accuracy and efficiency, in comparison with the existed typical location recommendation schemes.",
keywords = "Collaborative filtering, Community detection, Data fusion, Location recommendation",
author = "Wenjie Cai and Yufeng Wang and Ruheng Lv and Qun Jin",
year = "2019",
month = "7",
day = "1",
doi = "10.1016/j.compeleceng.2019.06.006",
language = "English",
volume = "77",
pages = "289--299",
journal = "Computers and Electrical Engineering",
issn = "0045-7906",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - An efficient location recommendation scheme based on clustering and data fusion

AU - Cai, Wenjie

AU - Wang, Yufeng

AU - Lv, Ruheng

AU - Jin, Qun

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Recently, location-based social networks (LBSNs) have witnessed a significant development, and have attracted millions of mobile users to share their locations and location-related contents. Due to the huge data in LBSNs, it is imperative to design an efficient location recommendation scheme for abundant users. But most of existing works suffer from poor accuracy and efficiency. On one hand, the traditional user-based collaborative filtering (CF) methods only focus on user characteristics, which limit the recommendation accuracy. On the other hand, it is inefficient to traverse all LBSN data in each recommendation. To solve the above issues, this paper proposes a new location recommendation method LC–G–P, in which, Louvain method is adopted to cluster users into several communities; and meanwhile, multiple features including Geographical distance, location Popularity are incorporated into location recommendation. Experiments based on real LBSN dataset illustrate that the proposed recommendation scheme LC–G–P has better accuracy and efficiency, in comparison with the existed typical location recommendation schemes.

AB - Recently, location-based social networks (LBSNs) have witnessed a significant development, and have attracted millions of mobile users to share their locations and location-related contents. Due to the huge data in LBSNs, it is imperative to design an efficient location recommendation scheme for abundant users. But most of existing works suffer from poor accuracy and efficiency. On one hand, the traditional user-based collaborative filtering (CF) methods only focus on user characteristics, which limit the recommendation accuracy. On the other hand, it is inefficient to traverse all LBSN data in each recommendation. To solve the above issues, this paper proposes a new location recommendation method LC–G–P, in which, Louvain method is adopted to cluster users into several communities; and meanwhile, multiple features including Geographical distance, location Popularity are incorporated into location recommendation. Experiments based on real LBSN dataset illustrate that the proposed recommendation scheme LC–G–P has better accuracy and efficiency, in comparison with the existed typical location recommendation schemes.

KW - Collaborative filtering

KW - Community detection

KW - Data fusion

KW - Location recommendation

UR - http://www.scopus.com/inward/record.url?scp=85067391744&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067391744&partnerID=8YFLogxK

U2 - 10.1016/j.compeleceng.2019.06.006

DO - 10.1016/j.compeleceng.2019.06.006

M3 - Article

VL - 77

SP - 289

EP - 299

JO - Computers and Electrical Engineering

JF - Computers and Electrical Engineering

SN - 0045-7906

ER -