Design And Implementation Of Agricultural Product E-commerce System Based On Deep Recommendatio

Agriculture is the primary industry in our country,and the sale of basic agricultural products directly affects farmers’ income and agricultural development.It is closely related to achieving common prosperity for the nation.With the development of information technology,agricultural e-commerce has brought new ideas and business opportunities to agricultural product sales.Recommendation systems are based on the historical interaction information between users and items or the feature information of users and items to meet users’ diverse and personalized shopping needs.However,in the context of agricultural e-commerce recommendations,due to the inability to accurately model users’ contextual semantic features,it becomes challenging to understand users’ intentions and achieve precise recommendations.To address the aforementioned issues,this article has accomplished the following two main tasks:This article proposes a deep learning recommendation model based on a fusion of TextRank and Bi-LSTM + Attention,which is utilized for user interest modeling and item feature modeling.Specifically,to address the issue of Bi-LSTM + Attention models’ inability to accurately extract features from excessively long texts,TextRank is introduced to extract key phrases from lengthy texts.Subsequently,a deep recommendation algorithm based on the fusion of TextRank and Bi-LSTM + Attention is proposed.In order to validate the effectiveness and performance of this approach for semantic recommendation,experiments are conducted on the Adressa dataset.The experimental results demonstrate that this recommendation method outperforms traditional approaches by better understanding users’ latent semantic information and providing more precise recommendations.Agricultural e-commerce system incorporating the recommendation algorithm based on TextRank fusion with Bi-LSTM + Attention has been designed and implemented.The e-commerce system is built on a B/S architecture and utilizes the Django development framework,My SQL database management system,and Redis.It includes user front-end shopping features and back-end management capabilities.The system’s stability has been verified through software testing,confirming that the system is fully implemented and effective.