At present, the fusion representation for multi-source texts is relatively simple, the difference between long and short texts is not considered, and the representation accuracy needs to be improved; in addition, when performing heterogeneous data fusion, the deep learning proposed in recent years can map each structural data to the same shared space. However, few studies have focused on user-generated content in e-commerce platforms. Therefore, we did a study on multi-source heterogeneous data fusion and representation strategies for user-generated content on e-commerce platforms. The convolutional neural network model is used to realize the fusion representation of heterogeneous data so that the various modalities and information of user-generated data can be considered when the product feature representation is performed, especially when the product text data is small, the integrated heterogeneous data may be it plays a better role in feature expansion, and further improves the accuracy and robustness of product feature representation in e-commerce platforms. A user preference estimation algorithm based on RBM is constructed in combination with the category attributes of the product itself; based on the existing explicit preference combined with the user's implicit preference, joint learning is performed to complete the user personalized recommendation based on collaborative filtering; the proposed algorithm is applied to multiple Amazon sub-datasets to verify the superiority of the proposed algorithm and the feasibility and accuracy of user-generated multi-source heterogeneous data fusion. The results show that the fusion of user-generated multi-source heterogeneous data can effectively improve the overall performance of the recommendation algorithm.

Xiaoyan Wang. Multi-source heterogeneous data fusion of cross-border e-commerce platform based on the neural network model. Academic Journal of Business & Management (2024) Vol. 6, Issue 1: 63-72. https://doi.org/10.25236/AJBM.2024.060109.

[1] Ranfagni S, Faraoni M, Zollo L, etal. Combining online market research methods for investigating brand alignment: the case of Nespresso [J]. British Food Journal, 2021, 123(13): 37-58.

[2] Wu X, Chen H, Wu G, etal. Knowledge Engineering with Big Data [J]. IEEE Intelligent Systems, 2015, 30(5): 1.

[3] Dong J, Zhuang D, Huang Y, etal. Advances in multi-sensor data fusion: algorithms and applications [J]. Sensors (Basel, Switzerland), 2009, 9(10): 7771-7784. 

[4] Burt P, Adelson E. The Laplacian Pyramid as a Compact Image Code [J]. Readings in Computer Vision, 1987, 31(4): 671-679.

[5] TOET A. Image fusion by a ratio of low-pass pyramid [J]. Pattern Recognition Letters, 1989, 9(4): 245-253.

[6] Marshall S, Matsopoulos G, Brunt J. Multiresolution morphological fusion of MR and CT images of the human brain [J]. IEE Proceedings - Vision, Image, and Signal Processing, 1994, 141(3): 137-142.

[7] Mitianoudis N, Stathaki T. Pixel-based and region-based image fusion schemes using. ICA bases[J]. Information Fusion, 2007, 8(2): 131-142.

[8] Liang J, Yang H, Ding L, etal. Image Fusion Using Higher Order Singular Value Decomposition [J]. IEEE Trans Image Process, 2012, 21(5): 2898-2909.

[9] Li S, Kwok J, Wang Y. Combination of images with diverse focuses using the spatial frequency [J]. Information Fusion, 2001, 2(3): 169-176.

[10] AYL, ASL, BZWA.Multi-focus image fusion with dense SIFT-Science Direct [J]. Information Fusion, 2015, 23(6): 139-155.

[11] Nejati M, Samavi S, Karimi N, Etal. Surface area-based focus criterion for multi-focusimage fusion [J]. Information Fusion, 2017, 36(5): 284-295.

[12] Jian L, Yang X, Zhou Z, etal. Multi-scale image fusion through rolling guidance filter [J]. Future Generation Computer Systems, 2018, 83(9): 310-325.

[13] Liu Y, Chen X, Peng H, etal. Multi-focus image fusion with a deep convolutional neural network k [J]. Information Fusion, 2017, 36(4): 191-207.

[14] Palsson F, Sveinsson J, Ulfarsson M. Multispectral and Hyperspectral Image Fusion Using a 3-D-Convolutional Neural Network [J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(5): 639-643.

[15] Richa, Bedi P. Trust and Distrust based Cross-domain Recommender System [J]. APPLIED Artificial Intelligence, 2021, 35(4): 326-351.

[16] Hardoon D R, Szedmak S, Shawe-Taylor J. Canonical correlation analysis; An overview with application to learning methods[J]. Neural Computation, 2004, 16(12): 2639-2664.

[17] Salakhutdinov R, Hinton G. An Efficient Learning Procedure for Deep Boltzmann Machines [J]. Neural Computation, 2012, 24(8): 1967-2006.

[18] Landgrebe Tcm, Duin, Rpw, Efficient Multiclass ROC Approximation by Decomposition via Confusion Matrix Perturbation Analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(5): 810-822.

[19] Salton G, Wong A. A Vector Space Model for Automatic Indexing [J]. Communications of the ACM, 1975, 18(11): 613-620.