Self-supervised representation learning for trip recommendation
Qiang Gao,Wei Wang,Kunpeng Zhang,Xin Yang,Congcong Miao,Tianrui Li
DOI: https://doi.org/10.1016/j.knosys.2022.108791
2022-04-01
Abstract:Trip recommendation is a significant and engaging location-based service that can help new tourists make more customized travel plans. It often attempts to suggest a sequence of points of interest (POIs) for a user who requests a personalized travel demand. Conventional methods either leverage the heuristic algorithms (e.g., dynamic programming) or statistical analysis (e.g., Markov models) to search or rank a POI sequence. These procedures may fail to capture the diversity of human needs and transitional regularities. They even provide recommendations that deviate from tourists' real travel intention when the trip data is sparse. Although recent deep recursive models (e.g., RNN) are capable of alleviating these concerns, existing solutions hardly recognize the practical reality, such as the diversity of tourist demands, uncertainties in the trip generation, and the complex visiting preference. Inspired by the advance in deep learning, we introduce a novel self-supervised representation learning framework for trip recommendation – SelfTrip, aiming at tackling the aforementioned challenges. Specifically, we propose a two-step contrastive learning mechanism concerning the POI representation, as well as trip representation. Furthermore, we present four trip augmentation methods to capture the visiting uncertainties in trip planning. We evaluate our SelfTrip on four real-world datasets, and extensive results demonstrate the promising gain compared with several cutting-edge benchmarks, e.g., up to 4% and 10% improvements on Osaka regarding F1 and pair-F1.
computer science, artificial intelligence