Recommender System in E-Commerce

Abstract

In the realm of e-commerce, recommendation systems play a pivotal role in guiding users towards relevant products. However, existing systems often grapple with inefficiencies in handling large datasets and fail to deliver personalized recommendations tailored to individual preferences. Addressing these challenges, the study introduces an innovative approach leveraging graph databases to enhance the performance of e- commerce recommendation systems. Through comprehensive analysis, the study delves into four critical aspects: database comparison, user exploration frequency across product categories, the diversity of available category types, and user browsing history analysis. This investigation unveils Neo4j's superior efficiency over MySQL in managing extensive datasets, laying the groundwork for more robust recommendation engines. By scrutinizing user behaviour patterns, the recommender system predicts preferences with precision, promising a tailored and gratifying shopping experience for users. Moreover, with support for a diverse array of category types, users gain flexibility in exploring products based on varied criteria, addressing a crucial need in the market for personalized shopping experiences. Leveraging insights gleaned from user browsing history, the system delivers refined recommendations, poised to elevate user satisfaction and engagement within the competitive landscape of e- commerce. In conclusion, the study highlights the significance of recommendation systems in enhancing the e- commerce experience. By leveraging graph databases, particularly Neo4j, over traditional systems like MySQL, significant improvements in managing extensive datasets are demonstrated.