Next POI recommendation with an adaptive gating mechanism embedded in graph neural networks

doi:10.13474/j.cnki.11-2246.2025.0715

Abstract

Abstract: The recommendation of the next POI has attracted much attention in location-based social networks,aiming to accurately recommend POIs based on users' historical check-ins and temporal information.However,traditional methods fail to consider temporal sequences and have low learning efficiency for graph nodes.To address this,in this paper,we propose a method for next POI recommendation by incorporating adaptive gating mechanisms into geographic and sequential graph modules.The proposed network consists of three main parts: the adaptive geographic graph module,which combines the adaptive gating mechanism with graph convolutional neural networks to adjust node fusion update weights by using gating signals; the adaptive sequential graph module,which learns user access preferences through a random walk network and enhances the weights of relevant preferences on the basis of target task attributes using adaptive gating mechanisms; and the semantic joint module,which maximizes the consistency distribution between the geographic and sequential graph modules and optimizes the loss of the joint framework via soft-label cross-entropy loss functions.To validate the effectiveness of the model,experiments are conducted on foreign datasets (Foursquare_NYC and Foursquare_TKY)and a domestic dataset(Microblog).The experimental results demonstrate that the proposed model achieves recommendation accuracies of over 85% across all datasets,with performance improvements ranging from 2.97% to 86.90% over those of state-of-the-art baseline models.

Key words: adaptive gating mechanism, next POI recommendation, graph neural networks

CLC Number:

P208

CHI Jinzhe, LIU Jiping, XU Shenghua, WANG Yong, WANG Zhuolu. Next POI recommendation with an adaptive gating mechanism embedded in graph neural networks[J]. Bulletin of Surveying and Mapping, 2025, 0(7): 90-96.

References

[1] REN Yueqiang,WANG Ze,SUN Xiaona,et al.A multi-element hybrid location recommendation algorithm for location based social networks[J].IEEE Access,2019,7:100416-100427.
[2] XIONG Xi,QIAO Shaojie,LI Yuanyuan,et al.A point-of-interest suggestion algorithm in Multi-source geo-social networks[J].Engineering Applications of Artificial Intelligence,2020,88:103374.
[3] SU Yijun,LI Xiang,LIU Baoping,et al.FGCRec:fine-grained geographical characteristics modeling for point-of-interest recommendation[C]//Proceedings of 2020 IEEE International Conference on Communications (ICC).Dublin,Ireland:IEEE,2020:1-6.
[4] WANG Jie,HUANG Zhiyi,LIU Zhaobin.SQPMF:successive point of interest recommendation system based on probability matrix factorization[J].Applied Intelligence,2024,54(1):680-700.
[5] XUN Yaling,WANG Yujia,ZHANG Jifu,et al.Higher-order embedded learning for heterogeneous information networks and adaptive POI recommendation[J].Information Processing & Management,2024,61(4):103763.
[6] DAI Shaojie,YU Yanwei,FAN Hao,et al.Spatio-temporal representation learning with social Tie for personalized POI recommendation[J].Data Science and Engineering,2022,7(1):44-56.
[7] BAO Yi,HUANG Zhou,LI Linna,et al.A BiLSTM-CNN model for predicting users' next locations based on geotagged social media[J].International Journal of Geographical Information Science,2021,35(4):639-660.
[8] ZHAO Pengpeng,LUO Anjing,LIU Yanchi,et al.Where to go next:a spatio-temporal gated network for next POI recommendation[J].IEEE Transactions on Knowledge and Data Engineering,2022,34(5):2512-2524.
[9] NIKOLENTZOS G,VAZIRGIANNIS M.Random walk graph neural networks[J].Advances in Neural Information Processing Systems,2020,33:16211-16222.
[10] ST JOHN P C,PHILLIPS C,KEMPER T W,et al.Message-passing neural networks for high-throughput polymer screening[J].The Journal of Chemical Physics,2019,150(23):234111.
[11] ZHANG Yufeng,YU Xueli,CUI Zeyu,et al.Every document owns its structure:inductive text classification via graph neural networks[EB/OL].[2024-10-08].https://arxiv.org/abs/2004.13826v2.
[12] JU Wei,QIN Yifang,QIAO Ziyue,et al.Kernel-based substructure exploration for next POI recommendation[C]//Proceedings of 2022 IEEE International Conference on Data Mining (ICDM).Orlando,FL:IEEE,2022:221-230.
[13] QIN Yifang,WU Hongjun,JU Wei,et al.A diffusion model for POI recommendation[J].ACM Transactions on Information Systems,2024,42(2):1-27.
[14] LUO Yingtao,LIU Qiang,LIU Zhaocheng.STAN:spatio-temporal attention network for next location recommendation[C]//Proceedings of 2021 Web Conference 2021.Ljubljana Slovenia:ACM Press,2021:2177-2185.
[15] YANG Song,LIU Jiamou,ZHAO Kaiqi.GETNext:trajectory flow map enhanced transformer for next POI recommendation[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval.Madrid Spain:ACM Press,2022:1144-1153.
[16] WU Yuxia,LI Ke,ZHAO Guoshuai,et al.Personalized long-and short-term preference learning for next POI recommendation[J].IEEE Transactions on Knowledge and Data Engineering,2022,34(4):1944-1957.
[17] SUN Ke,QIAN Tieyun,CHEN Tong,et al.Where to go next:modeling long-and short-term user preferences for point-of-interest recommendation[J].Proceedings of the AAAI Conference on Artificial Intelligence,2020,34(1):214-221.
[18] QIN Yifang,WANG Yifan,SUN Fang,et al.DisenPOI:disentangling sequential and geographical influence for point-of-interest recommendation[C]//Proceedings of the 16th ACM International Conference on Web Search and Data Mining.Singapore Singapore:ACM Press,2023:508-516.