Enhanced Predictive E-Commerce Model for Online Auction Using Convolutional Neural Network and Long Short Term Memory

Authors

  • Miriam Ayanaye Stephen Ignatius Ajuru Unversity of Education
  • Promise A. Nlerum Ignatius Ajuru Unversity of Education
  • Henry O. Ordu Ignatius Ajuru Unversity of Education
  • Emmanuel J. Izionworu Ignatius Ajuru Unversity of Education

DOI:

https://doi.org/10.55123/jomlai.v5i3.8958

Keywords:

Online Auctions, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Price Prediction, Fraud Detection

Abstract

This study modelled an enhanced predictive e-commerce platform for online auctions in B2C and C2C markets, leveraging a hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model to improve transaction efficiency and reliability. The system addresses key auction challenges, including price uncertainty, fraudulent behaviors, lack of user trust, and information overload. By processing multi-modal data bid histories, item images, and metadata the CNN extracts critical features like bid patterns and visual cues, The  hybrid CNN-LSTM model achieved approximately 90% accuracy in auction price prediction, representing a substantial improvement over the approximately 70% accuracy of the existing rule-based system. The model also reduced the Mean Absolute Percentage Error (MAPE) from approximately 15% to 5%, indicating more reliable estimation of final auction prices and reduced prediction error. For fraudulent-bid detection, the proposed model achieved approximately 85% precision, compared with 60% for the existing system, demonstrating improved ability to distinguish potentially fraudulent bidding activities while reducing the risk of misleading alerts. Beyond predictive accuracy, the model improved processing efficiency, decision support, and item-authenticity validation by jointly analyzing bid histories, item images, and auction metadata. The principal novelty of the study lies in the integration of CNN-based spatial/visual feature extraction, bidirectional LSTM-based temporal sequence modelling, and attention mechanisms within a multi-modal auction framework. Unlike previous approaches that primarily rely on single-modal data or address price forecasting and fraud detection separately, the proposed architecture simultaneously captures visual cues, bidding dynamics, and contextual auction information to support price prediction, fraud detection, authenticity verification, and personalized bidding recommendations. These results demonstrate that the proposed model provides a more comprehensive and context-aware approach to improving reliability, efficiency, and decision-making in B2C and C2C online auctions.

References

[1] A. Bhatia, R. Sharma, and S. Patel, "Advanced LSTM models for e-commerce inventory forecasting," J. Bus. Anal., vol. 7, no. 3, pp. 123–135, 2024. doi: 10.1016/j.jba.2024.01.005.

[2] O. Sushmita, L. Oloti, and O. Fatimah, "Soft computing framework," Int. J. Comput. Appl., vol. 4, no. 11, pp. 21–29, 2020.

[3] B. Lahm, U. Mozi, and I. Gebb, "E-commerce systems," in Proc. 17th Int. Conf. Inf. Syst., Cleveland, OH, 2020, pp. 381–390.

[4] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Comput., vol. 9, no. 8, pp. 1735–1780, 1997. doi: 10.1162/neco.1997.9.8.1735.

[5] M. Sadia, A. Mahfujul, and T. Kamrul, "Age estimation from facial image using convolutional neural network (CNN)," in Int. Conf. Comput., Commun., Chem., Mater. Electron. Eng. (IC4ME2), IEEE, 2019.

[6] X. Wu and J. Chen, "Enhanced CNN-LSTM for auction price interval prediction," in IEEE Int. Conf. Artif. Intell. Appl. (AIAA), 2024, pp. 67–74. doi: 10.1109/AIAA.2024.7890123.

[7] J. Chen and X. Wu, "LSTM-based price interval prediction for online auctions," IEEE Trans. Artif. Intell., vol. 4, no. 2, pp. 89–97, 2023. doi: 10.1109/TAI.2023.1234567.

[8] I. Pandev, "T-commerce – What the tablet can do for brands and their consumers," Governor Technology, 2020. [Online]. Available: https://www.governor.co.uk/t-commerce/ (Accessed: May 4, 2020).

[9] S. Mohammed, C. Soumya, P. Sravani, and P. Tejashwini, "A novel approach to e-commerce interface," Int. J. Creative Res. Thoughts (IJCRT), 2023. [Online]. Available: https://ijcrt.org/papers/IJCRT2305279.pdf

[10] T. Marcus, L. Operio, and W. Tayo, "G-CORE: A core for future graph intelligence gathering languages," Inf. Syst. Res., vol. 4, no. 2, pp. 166–204, 2021.

[11] P. Machala, U. Sondos, and M. Blessing, "Criminal face detection system: Using the Fisherface algorithm and the Local Binary Patterns Histogram (LBPH) classifier," in 7th Int. Annu. Eng. Seminar, Yogyakarta, 2023.

[12] P. Juyan, "Some studies in machine learning using the game of checkers," IBM J. Res. Dev., vol. 3, no. 3, pp. 210–229, 2021. doi: 10.1147/rd.33.0210.

[13] H. Isak, O. Kemi, and U. Linga, "Voronoi-based multi-robot autonomous exploration in unknown environments via deep reinforcement learning," IEEE Trans. Veh. Technol., vol. 69, no. 12, pp. 14413–14423, 2022. doi: 10.1109/tvt.2022.3034800.

[14] T. Ryosuke, H. Jinsong, and P. Wang, "Photo-real talking head with deep bidirectional LSTM," in Proc. ICASSP 2022, IEEE, 2022, pp. 1–5.

[15] J. Chen and F. Liu, "Hybrid CNN-LSTM for supply chain demand forecasting in cross-border e-commerce," J. Supply Chain Manag., vol. 60, no. 2, pp. 45–58, 2024. doi: 10.1111/jscm.12389.

[16] X. Wang, L. Chen, and T. Zhao, "Bidirectional LSTM for volatile auction markets," in IEEE Int. Conf. Mach. Learn. Appl. (ICMLA), 2025, pp. 45–52. doi: 10.1109/ICMLA.2025.7891234.

[17] H. Zhang and X. Wang, "ARIMA-CNN-LSTM for retail sales prediction," J. Bus. Anal., vol. 6, no. 4, pp. 156–168, 2023. doi: 10.1016/j.jba.2023.02.007.

[18] R. Zhao, W. Yang, and S. Kim, "Interpretable CNN-LSTM for cryptocurrency price forecasting," arXiv, 2023. [Online]. Available: https://arxiv.org/abs/2302.07890

[19] H. Zhang and X. Wang, "ARIMA-CNN-LSTM for retail sales prediction," J. Bus. Anal., vol. 6, no. 4, pp. 156–168, 2023. doi: 10.1016/j.jba.2023.02.007.

[20] R. Zhao, W. Yang, and S. Kim, "Interpretable CNN-LSTM for cryptocurrency price forecasting," arXiv, 2023. [Online]. Available: https://arxiv.org/abs/2302.07890

[21] S. Patel, M. Gupta, and A. Rao, "Multi-modal CNN-LSTM for fraud detection in C2C auctions," Comput. Secur., vol. 135, p. 104012, 2025. doi: 10.1016/j.cose.2025.104012.

Downloads

Published

2026-09-29

How to Cite

Stephen, M. A., Promise A. Nlerum, Henry O. Ordu, & Emmanuel J. Izionworu. (2026). Enhanced Predictive E-Commerce Model for Online Auction Using Convolutional Neural Network and Long Short Term Memory. JOMLAI: Journal of Machine Learning and Artificial Intelligence, 5(3), 122–131. https://doi.org/10.55123/jomlai.v5i3.8958

Issue

Vol. 5 No. 3 (2026): September 2026

Section

Articles

License

Copyright (c) 2026 Miriam Ayanaye Stephen, Promise A. Nlerum, Henry O. Ordu, Emmanuel J. Izionworu

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Copyright (c) 2022 The authors. Published by Yayasan Literasi Indonesia
Creative Commons License This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License

The author(s) whose article is published in the JOMLAI journal attain the copyright for their article and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. By submitting the manuscript to JOMLAI, the author(s) agree with this policy. No special document approval is required.

The author(s) guarantee that:their article is original, written by the mentioned author(s),

  • has never been published before,
  • does not contain statements that violate the law, and
  • does not violate the rights of others, is subject to copyright held exclusively by the author(s), and is free from the rights of third parties, and that the necessary written permission to quote from other sources has been obtained by the author(s).

The author(s) retain all rights to the published work, such as (but not limited to) the following rights:

  • Copyright and other proprietary rights related to the article, such as patents,
  • The right to use the substance of the article in its own future works, including lectures and books,
  • The right to reproduce the article for its own purposes,
  • The right to archive all versions of the article in any repository, and
  • The right to enter into separate additional contractual arrangements for the non-exclusive distribution of published versions of the article (for example, posting them to institutional repositories or publishing them in a book), acknowledging its initial publication in this journal (JOMLAI: Journal of Machine Learning and Artificial Intelligence).

Suppose the article was prepared jointly by more than one author. Each author submitting the manuscript warrants that all co-authors have given their permission to agree to copyright and license notices (agreements) on their behalf and notify co-authors of the terms of this policy. JOMLAI will not be held responsible for anything that may arise because of the writer's internal dispute. JOMLAI will only communicate with correspondence authors.

Authors should also understand that their articles (and any additional files, including data sets, and analysis/computation data) will become publicly available once published. The license of published articles (and additional data) will be governed by a Creative Commons Attribution-ShareAlike 4.0 International License. JOMLAI allows users to copy, distribute, display and perform work under license. Users need to attribute the author(s) and JOMLAI to distribute works in journals and other publication media. Unless otherwise stated, the author(s) is a public entity as soon as the article is published