The current product recommendation system has the problem of overly relying on user historical behavior data and ignoring the complex interrelationships between products. Given this, this paper proposes a vertical e-commerce product recommendation method based on an improved graph neural network. This method strengthens the graph neural network’s ability to capture detailed product and user preference relationships. Information transmission mechanism is introduced, and reinforcement learning techniques are integrated to improve the model’s recommendation accuracy and adaptability. Experimental results showed that the model achieved an accuracy rate of 0.89 on the FITB task of the POG dataset on the Alibaba platform, representing a 20% improvement over other models. Additionally, on the Amazon Reviews dataset, the model achieved an R2 of 0.81, demonstrating strong generalization capabilities. The results indicate that the proposed method offers significant advantages in improving the accuracy and personalization of e-commerce recommendation systems. This study provides a new theoretical foundation and practical guidance for vertical product recommendation systems in e-commerce.
FelixARembulanGD. Analysis of key factors for improved customer experience, engagement, and loyalty in the E-Commerce industry in Indonesia. Aptisi Transactions on Technopreneurship (ATT)2023; 5(2): 196–208.
2.
KarimKSIslamMEIbrahimAM, et al.Online marketing trends and purchasing intent: advances in customer satisfaction through PLS-SEM and ANN approach. Advances in Decision Sciences2023; 27(4): 1–31.
3.
GaoCZhengYLiN, et al.A survey of graph neural networks for recommender systems: challenges, methods, and directions. ACM Trans Recomm Syst2023; 1(1): 1–51.
4.
FkihF. Similarity measures for collaborative filtering-based recommender systems: review and experimental comparison. J King Saud Univ Comput Inf Sci2022; 34(9): 7645–7669.
5.
GaoCZhengYWangW, et al.Causal inference in recommender systems: a survey and future directions. ACM Trans Inf Syst2024; 42(4): 1–32.
6.
LiuHZhengCLiD, et al.EDMF: efficient deep matrix factorization with review feature learning for industrial recommender system. IEEE Trans Industr Inform2021; 18(7): 4361–4371.
7.
LiDLiuHZhangZ, et al.CARM: confidence-aware recommender model via review representation learning and historical rating behavior in the online platforms. Neurocomputing2021; 455: 283–296.
8.
NasirMEzeifeCI. A survey and taxonomy of sequential recommender systems for e-commerce product recommendation. SN Computer Science2023; 4(6): 708–740.
9.
XuJHuZZouJ. Personalized product recommendation method for analyzing user behavior using DeepFM. Journal of Information Processing Systems2021; 17(2): 369–384.
10.
DengW. Leveraging consumer behaviors for product recommendation: an approach based on heterogeneous network. Electron Commer Res2022; 22(4): 1079–1105.
11.
NasirMEzeifeCI. Semantic enhanced Markov model for sequential E-commerce product recommendation. Int J Data Sci Anal2023; 15(1): 67–91.
12.
AtinaVHartantiD. Knowledge-based recommendation modeling for clothing product selection recommendation system. J Tek Inform2022; 3(5): 1407–1413.
13.
ZhangWChenZZhaH, et al.Learning from substitutable and complementary relations for graph-based sequential product recommendation. ACM Trans Inf Syst2021; 40(2): 1–28.
14.
LiDDengCWangX, et al.Joint inter-word and inter-sentence multi-relation modeling for summary-based recommender system. Inf Process Manag2024; 61(3): 103631.
15.
LiuBLiDWangJ, et al.Integrating user short-term intentions and long-term preferences in heterogeneous hypergraph networks for sequential recommendation. Inf Process Manag2024; 61(3): 103680.
16.
LiuHZhengCLiD, et al.Multi-perspective social recommendation method with graph representation learning. Neurocomputing2022; 468: 469–481.
17.
JiangWLuoJHeM, et al.Graph neural network for traffic forecasting: the research progress. ISPRS Int J GeoInf2023; 12(3): 100–134.
18.
YangXYanMPanS, et al.Simple and efficient heterogeneous graph neural network. Proc AAAI Conf Artif Intell2023; 37(9): 10816–10824.
19.
ChoudharyKDeCostB. Atomistic line graph neural network for improved materials property predictions. npj Comput Mater2021; 7(1): 185–192.
20.
ZengYTangJ. RLC-GNN: an improved deep architecture for spatial-based graph neural network with application to fraud detection. Appl Sci2021; 11(12): 5656.
21.
ZhengCChenHChengY, et al.ByteGNN: efficient graph neural network training at large scale. Proceedings VLDB Endowment2022; 15(6): 1228–1242.
22.
XuejianHYuyangLTinghuaiM. Classification model for scholarly articles based on improved graph neural network. Data Analysis and Knowledge Discovery2022; 6(10): 93–102.
23.
GuptaAMattaPPantB. Graph neural network: current state of art, challenges and applications. Mater Today Proc2021; 46(20): 10927–10932.
24.
YangHYaoLCaiJ, et al.A new interest extraction method based on multi-head attention mechanism for CTR prediction. Knowl Inf Syst2023; 65(8): 3337–3352.
25.
GuoKHuYSunY, et al.Hierarchical graph convolution network for traffic forecasting. Proc AAAI Conf Artif Intell2021; 35(1): 151–159.
26.
LiuRChangQWangY, et al.Feature entropy adaptive network for weak magnetic signal classification. IEEE Sens J2023; 24(2): 1322–1332.
27.
ChangQLiuRWangY, et al.High-precision magnetic field reconstruction and anomaly classification. IEEE Sens J2023; 23(17): 19163–19175.
28.
PanchbhaiKGLanjewarMGNaikAV. Modified MobileNet with leaky ReLU and LSTM with balancing technique to classify the soil types. Earth Sci Inform2025; 18(1): 1–22.
29.
PreethiPMamathaHR. Region-based convolutional neural network for segmenting text in epigraphical images. Artif Intell Appl2023; 1(2): 119–127.
30.
SunPYTAndersonMHGangHI. Determining the hierarchical structure and nature of servant leadership. J Bus Psychol2024; 39(3): 715–734.
31.
ForestFLebbahMAzzagH, et al.Deep embedded self-organizing maps for joint representation learning and topology-preserving clustering. Neural Comput Appl2021; 33(24): 17439–17469.
32.
LiuHLiLYuN, et al.Outfit compatibility model using fully connected self-adjusting graph neural network. Vis Comput2024; 40(11): 8331–8343.
33.
YangXSongXFengF, et al.Attribute-wise explainable fashion compatibility modeling. ACM Trans Multimedia Comput Commun Appl2021; 17(1): 1–21.
34.
ShiWAnWTianF, et al.A unified knowledge transfer network for generalized category discovery. Proc AAAI Conf Artif Intell2024; 38(17): 18961–18969.
35.
XuNGooiHBWangL, et al.Loop optimization noise-reduced LSTM based classifier for PD detection. IEEE Trans Ind Appl2022; 59(1): 392–402.
36.
LiYShenJMaoZ. CoocNet: a novel approach to multi-label text classification with improved label co-occurrence modeling. Appl Intell2024; 54(17): 8702–8718.
