A POI recommendation approach integrating social spatio-temporal information into probabilistic matrix factorization
- PDF / 879,195 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 35 Downloads / 205 Views
A POI recommendation approach integrating social spatio-temporal information into probabilistic matrix factorization Mehri Davtalab1 · Ali Asghar Alesheikh1 Received: 27 May 2019 / Revised: 19 August 2020 / Accepted: 22 August 2020 © Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract In recent years, point of interest (POI) recommendation has gained increasing attention all over the world. POI recommendation plays an indispensable role in assisting people to find places they are likely to enjoy. The exploitation of POIs recommendation by existing models is inadequate due to implicit correlations among users and POIs and cold start problem. To overcome these problems, this work proposed a social spatio-temporal probabilistic matrix factorization (SSTPMF) model that exploits POI similarity and user similarity, which integrates different spaces including the social space, geographical space and POI category space in similarity modelling. In other words, this model proposes a multivariable inference approach for POI recommendation using latent similarity factors. The results obtained from two real data sets, Foursquare and Gowalla, show that taking POI correlation and user similarity into account can further improve recommendation performance. In addition, the experimental results show that the SSTPMF model performs better in alleviating the cold start problem than state-of-the-art methods in terms of normalized discount cumulative gain on both data sets. Keywords Point of interest · Location-based social network · Probabilistic matrix factorization
1 Introduction The recent prevalence of mobile devices and the development of position localization techniques have enhanced location-based social networking (LBSN) services, allowing users to share their locations and check-in information with others via interconnection relationships [1]. The growth of the public acceptance of LBSN services during the last decades has increased the need to provide personalized location-based services that are related to venues and points of interest (POIs). Recently, POI recommendation as location-based recommen-
B 1
Ali Asghar Alesheikh [email protected] Department of Geospatial Information Systems, Faculty of Geodesy and Geomatics Engineering, K. N. Toosi University of Technology, Valiasr Street, Mirdamad Cross, Tehran 19967-15433, Iran
123
M. Davtalab, A. A. Alesheikh
dation has attracted many researchers. Various POI recommendation approaches to identify relevant POIs from an overwhelming number of locations have been introduced [2–4]. In spite of the traditional recommendation, in LBSNs a user is able to choose where and when to check-in, so the POI recommendation is generally based on the users’ checkin data. When check-in data are used in POI recommendation, several challenges are also encountered, including the data sparsity issue and the cold start problem. Data sparsity occurs when the number of entries in the user-POI matrix, which includes the frequency of the POIs checked by users, is notably low [5]
Data Loading...
