TY - GEN
T1 - Inferring continuous latent preference on transition intervals for next point-of-interest recommendation
AU - He, Jing
AU - Li, Xin
AU - Liao, Lejian
AU - Wang, Mingzhong
N1 - Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - Temporal information plays an important role in Point-of-Interest (POI) recommendations. Most existing studies model the temporal influence by utilizing the observed check-in time stamps explicitly. With the conjecture that transition intervals between successive check-ins may carry more information for diversified behavior patterns, we propose a probabilistic factor analysis model to incorporate three components, namely, personal preference, distance preference, and transition interval preference. They are modeled by an observed third-rank transition tensor, a distance constraint, and a continuous latent variable, respectively. In our framework, the POI recommendation and the transition interval for user’s very next move can be inferred simultaneously by maximizing the posterior probability of the overall transitions. Expectation Maximization (EM) algorithm is used to tune the model parameters. We demonstrate that the proposed methodology achieves substantial gains in terms of prediction on next move and its expected time over state-of-the-art baselines. Code related to this paper is available at: https://github.com/skyhejing/ECML-PKDD-2018.
AB - Temporal information plays an important role in Point-of-Interest (POI) recommendations. Most existing studies model the temporal influence by utilizing the observed check-in time stamps explicitly. With the conjecture that transition intervals between successive check-ins may carry more information for diversified behavior patterns, we propose a probabilistic factor analysis model to incorporate three components, namely, personal preference, distance preference, and transition interval preference. They are modeled by an observed third-rank transition tensor, a distance constraint, and a continuous latent variable, respectively. In our framework, the POI recommendation and the transition interval for user’s very next move can be inferred simultaneously by maximizing the posterior probability of the overall transitions. Expectation Maximization (EM) algorithm is used to tune the model parameters. We demonstrate that the proposed methodology achieves substantial gains in terms of prediction on next move and its expected time over state-of-the-art baselines. Code related to this paper is available at: https://github.com/skyhejing/ECML-PKDD-2018.
KW - Point-of-Interest
KW - Probabilistic factor analysis model
KW - Recommendation
UR - http://www.scopus.com/inward/record.url?scp=85061156359&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-10928-8_44
DO - 10.1007/978-3-030-10928-8_44
M3 - Conference contribution
AN - SCOPUS:85061156359
SN - 9783030109271
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 741
EP - 756
BT - Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2018, Proceedings
A2 - Bonchi, Francesco
A2 - Berlingerio, Michele
A2 - Gärtner, Thomas
A2 - Hurley, Neil
A2 - Ifrim, Georgiana
PB - Springer Verlag
T2 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML-PKDD 2018
Y2 - 10 September 2018 through 14 September 2018
ER -