Blockchain for recommender systems

Blockchain for recommender systems

A dependable Recommender Systems (RS) is crucial in E-commerce. Worldwide vendors understand RS's importance in business growth. RS is highly favoured in E- commerce, and IoT. However, privacy concerns like shared devices and strangers accessing data hinder data sharing. Blockchain solutions seem promising to addressin addressing these issues due to their security, verifiability, robustness, and availability. Blockchain is part of distributed networks. The fusion of IoT and blockchain technology has created a modern distributed system. The integration guarantees the security and scalability of the recommender system. Combining the recommendation system and blockchain technology makes online activities safer and more private. Blockchain can consider RS’s characteristics, including transparency of transactions or recommendation models, the immutability of associated blocks, and the traceability of non-repudiation. This thesis is devoted to the a detailed understanding of building RS and addressing privacy solutions on thats. The main contributions of this thesis are related to the design and development of a privacy solution for RS using a blockchain based method, evaluation of the performance of a recommender system, and the prediction of such performance, where we have built a RS for “Movielens” data using blockchain based method and addressed several issues regarding both topics. In this study, we proposed a simple blockchain movie recommendation system. Our research provides a detailed understanding of RS’s Blockchain-related privacy resolutions and insight into present applications in various fields. The proposed technique is being assessed and tested by some users, and the results show that the proposed system protects MS privacy effectively. eExperiments on Movielens datasets have been compared with other algorithms. According to our experimental findings, DualLight - GCN achieves the best recommendation performance. The ablation study additionally demonstrates the innovation of the DualLight - GCN method in the multimodal fusion field. We also classified the movielens dataset using five standard classification methods and found that the MiFinn method outperformed both AUC and GAUC methods compared to similar practices. We intend to explore the application of different recommendation algorithms with other innovative contract platforms. Smart contracts, while secure, can be computationally intensive, especially when dealing with large datasets like Movielens. This can lead to increased processing time and higher energy consumption, affecting the overall system performance. While our work iii using blockchain offers significant advantages, scalability and computational overhead remain limitations. Future work includes improving the proposed project-based collaboration scheme and integrating procedure-related explainable recommender systems with blockchain analytics.

File reproduced in accordance with the copyright policy of the publisher/author/creator.