REVOLUTIONIZING TREND RECOMMENDATIONS: A DEEP LEARNING APPROACH FOR IMAGE-BASED INSIGHTS
DOI:
https://doi.org/10.29121/shodhkosh.v4.i1.2023.2859Keywords:
Deep Learning, Image-Based Recommendation, Convolutional Neural Networks (CNN), Feature Extraction, Image AnalysisAbstract [English]
The rapid increases in visual content creation across different sectors, starting from social media to even e-commerce, have had a major impact on their respective users, who want bespoke experiences: preferences are recognized and satisfied by recommendations contextualized within a specific sense of the world. Over many years, traditional recommendation methods took into consideration only textual content, user profiles, or purchase history as the material for recommendations. As images become the staple of engagement in the digital world, the old system misses those very nuances of visual preference that affect choice through color, texture, and style.
This shift has paved way for image-based recommendation systems. Indeed, these are further dependent on powerful machine learning and deep learning models in the analysis of visual data to discern meaningful patterns in drawing recommendations in line with a user's visual taste. In such industries as fashion and design, social media, style and aesthetic become much stronger drivers of engagement.
We propose in this paper a sophisticated AI-based Trend Recommendation System, which combines deep learning model feature extraction abilities with the precision of the Nearest Neighbor Search algorithms. Our system has at its core the ResNet50 model, a widely recognized CNN which is associated with superior ability to analyze images. This is achieved using ResNet50 to extract features from the images at a deep level so that their style and trend may be characterized with intricate visual characteristics. These features are then compared to a large curated dataset using Nearest Neighbor Search, in turn ensuring that the recommended images are not only visually relevant but also contextually accurate.
Our research expands ResNet50 by evaluating other state-of-the-art CNN models, including VGG16, InceptionV3, and MobileNetV2. These models are tested on their ability to extract features, provide recommendations with accuracy, and consume fewer computations. By comparing these models, we will know which one is more effective for real applications for speed and precision.This study is a landmark in developing an image-based recommender system and finds how deep learning has the transforming capability to make experience highly personalized and visually oriented.
References
A. van den Oord, S. Dieleman, and B. Schrauwen, "Deep content-based music recommendation," in Proceedings of the 26th International Conference on Neural Information Processing Systems (NIPS), 2013, pp. 2643-2651.
P.-S. Huang, X. He, J. Gao, L. Deng, A. Acero, and L. Heck, "Learning deep structured semantic models for web search using clickthrough data," in Proceedings of the 22nd ACM International Conference on Information and Knowledge Management (CIKM), 2013, pp. 2333-2338. DOI: https://doi.org/10.1145/2505515.2505665
H.-F. Wang, N. Wang, and D.-Y. Yeung, "Collaborative deep learning for recommender systems," in Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2015, pp. 1235-1244. DOI: https://doi.org/10.1145/2783258.2783273
S. Sedhain, A. K. Menon, S. Sanner, and L. Xie, "Autorec: Autoencoders meet collaborative filtering," in Proceedings of the 24th International World Wide Web Conference (WWW), 2015, pp. 111-112.USDA Animal and Plant Health Inspection Service https://www.aphis.usda.gov/plant-pests-diseases/citrus-diseases/citrus-greening-and-asian-citrus-psyllid#:~:text=Citrus%20greening%2C%20also%20called%20Huanglongbing,There%20is%20no%20cure.
M. Ali, M. A. Bacha, and S. Ahmad, "A multi-view deep learning approach for cross-domain user modeling in recommendation systems," in Proceedings of the 24th International World Wide Web Conference (WWW), 2015, pp. 601-602.
S. Li, J. Kawale, and Y. Fu, "Deep collaborative filtering via marginalized denoising auto-encoder," in Proceedings of the 22nd ACM International Conference on Information and Knowledge Management (CIKM), 2015, pp. 811-820. DOI: https://doi.org/10.1145/2806416.2806527
R. He and J. McAuley, "VBPR: Visual Bayesian personalized ranking from implicit feedback," in Proceedings of the 30th AAAI Conference on Artificial Intelligence (AAAI), 2016, pp. 144-150. DOI: https://doi.org/10.1609/aaai.v30i1.9973
B. Hidasi, A. Karatzoglou, L. Baltrunas, and D. Tikk, "Session-based recommendations with recurrent neural networks," in Proceedings of the 4th International Conference on Learning Representations (ICLR), 2016.
P. Covington, J. Adams, and E. Sargin, "Deep neural networks for YouTube recommendations," in Proceedings of the 10th ACM Conference on Recommender Systems (RecSys), 2016, pp. 191-198. DOI: https://doi.org/10.1145/2959100.2959190
H. Cheng, L. Koc, J. Harmsen, T. Shaked, T. Chandra, H. Aradhye, G. Anderson, G. Corrado, W. Chai, M. Ispir, R. Anil, Z. Haque, L. Hong, V. Jain, and X. Zhou, "Wide & deep learning for recommender systems," in Proceedings of the Workshop on Deep Learning for Recommender Systems at RecSys, 2016, pp. 7-10. DOI: https://doi.org/10.1145/2988450.2988454
Y. Zheng, B. Tang, W. Ding, and H. Zhou, "A neural autoregressive approach to collaborative filtering," in Proceedings of the 33rd International Conference on Machine Learning (ICML), 2016, pp. 764-773.
W. Wu, A. Ahmed, A. Beutel, E. Smola, and H. Jing, "Collaborative denoising auto-encoders for top-n recommender systems," in Proceedings of the 10th ACM International Conference on Web Search and Data Mining (WSDM), 2016, pp. 153-162. DOI: https://doi.org/10.1145/2835776.2835837
Y. Tan, X. Xu, L. Li, and W. Cheng, "Improved recurrent neural networks for session-based recommendations," in Proceedings of the Workshop on Deep Learning for Recommender Systems at RecSys, 2016, pp. 17-22. DOI: https://doi.org/10.1145/2988450.2988452
J. Lian, X. Zhou, F. Zhang, Z. Chen, X. Xie, and G. Sun, "CCCFNet: A content-boosted collaborative filtering neural network for cross-domain recommender systems," in Proceedings of the 26th International World Wide Web Conference (WWW), 2017, pp. 817-818. DOI: https://doi.org/10.1145/3041021.3054207
X. He, L. Liao, H. Zhang, L. Nie, X. Hu, and T.-S. Chua, "Neural collaborative filtering," in Proceedings of the 26th International World Wide Web Conference (WWW), 2017, pp. 173-182. DOI: https://doi.org/10.1145/3038912.3052569
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Ashlin Deepa R N, V. Srilakshmi, Ch. Vidyadhari, Srihithagunapriya Nimmala, Mallikarjuna Rao Gundarapu
This work is licensed under a Creative Commons Attribution 4.0 International License.
With the licence CC-BY, authors retain the copyright, allowing anyone to download, reuse, re-print, modify, distribute, and/or copy their contribution. The work must be properly attributed to its author.
It is not necessary to ask for further permission from the author or journal board.
This journal provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge.
