Enhancing Service Excellence: Blockchain-AI TF-IDF Recommendations
Abubakar Din Mohamed, Dong-Seong Kim, Tae-Soo Jun · 한국통신학회논문지 · 2024
Recommendation systems, ubiquitous across diverse sectors such as e-commerce, streaming services, and social media, play a pivotal role in tailoring user experiences. However, their application remains underexplored in sectors like dealerships and vehicles, where personalized suggestions can significantly enhance customer engagement and decision-making. Despite their widespread use, limited attention has been directed towards optimizing recommendation systems for the unique dynamics of the dealership and vehicle sectors, presenting an untapped potential for improvement and innovation. Utilizing software, artificial intelligence, and algorithms, our system addresses user complaints by seamlessly integrating AI algorithms and blockchain technology for enhanced security. Leveraging the Term Frequency-Inverse Document Frequency of Records (TF-IDF) vectorization for precision, the system demonstrates remarkable accuracy (99.8%) through cosine similarity (CS) and K-Nearest Neighbors evaluation. Propelled by advanced AI algorithms, it outperforms other blockchain-based recommendation systems, showcasing its potential in dealership and vehicle-related contexts.