RECOMMENDATION SYSTEMS IN E-COMMERCE PLATFORMS BASED ON GRAPH NEURAL NETWORKS
DOI:
https://doi.org/10.24867/27BE18MatovicKeywords:
graph neural networks, recommendation systems, e-commerce, matrix factorizationAbstract
In e-commerce, a large amount of money is invested annually in internet marketing. These investments aim to present products to a broader range of users so that they can be shown to more people interested in those products. A recommendation system could reduce these costs and increase profit by displaying personalized recommendations tailored to each user's taste. In this regard, this paper explores several approaches to implementing a recommendation system: a traditional approach and approaches based on the modern architecture of graph neural networks. Data on user interactions with items and data on users and items themselves were collected for learning. This paper aims to examine the quality of modern methods based on graph neural networks, compare them with traditional methods, and explore their underexplored advantages.
References
[2] F. Scarselli, M. Gori, A. C. Tsoi, M. Hagenbuchner и G. Monfardini, „The Graph Neural Network Model,“ IEEE Transactions on Neural Networks, т. I, бр. 20, pp. 61-80, 2008.
[3] M. Balabanovic и Y. Shoham, „Fab: content-based, collaborative recommendation,“ Communications of the ACM, т. III, бр. 40, pp. 66-72, 1997.
[4] Y. Koren, R. Bell и C. Voinsky, „Matrix tactorization techinques for recommender systems,“ Computer, т. 42, бр. 8, pp. 30-37, 2009.
[5] Y. Koren, „The Bellkor Solution to the Netflix Grand Prize,“ у Netlix Prize Documentation, т. 81, 2009, pp. 1-10.
[6] W. Hamilton, Z. Ying и J. Leskovec, „Inductive Representation Learning on Large Graphs,“ у Advances in Neural Infromation Processing Systems, 2017.
[7] X. Wang, X. He, M. Wang и F. Feng, „Neural Graph Collaborative Filtering,“ у 42nd international ACM SIGIR conference on Research and development in Information Retrieval, 2019.
[8] W. Xiao, H. Ji, C. Shi, B. Wang, Y. Ye и C. Peng, „Heterogeneous Graph Attention Network,“ у The world wide web conference, 2019.
[9] OList и A. Sionek, „OList Brazillian Ecommerce Dataset [Kaggle],“ [На мрежи]. Available: https://doi.org/10.34740/KAGGLE/DSV/195341. [Последњи приступ 12 July 2023].
[10] M. Davis и M. Palumbo, „A primer on the economics and time series econometrics of wealth effects,“ 2001.
[11] J. Leskovec, A. Rajaraman и J. Ullman D, Mining of Massive Datasets, Cambridge University Press, 2010.
[12] T. Kipf и M. Welling, „Semi-supervised Classification with Graph Convolutional Networks,“ arXiv, 2016.
[13] P. Velickovic, C. Guillem, A. Casanova и A. Romero, „Graph Attention Networks,“ arXiv, 2017.
