Precise Recommendation System for Agricultural Products in E-Commerce
To solve the problem of low precision of Individualized Recommendation of agricultural products in e-commerce, the Precise Recommendation System for Agricultural Products in E-commerce is researched. Aiming at the insufficiency of the classic Apriori algorithm, a new weighted fuzzy association rules mining algorithm is put forward to ensure the downward closure of frequent itemsets. The workflow of the recommendation system was tested through the structural design of e⁃commerce recommendation system, data preprocessing module design and recommendation module design. The hit rate is selected as the evaluation standard of different recommendation models. The contrastive analysis for the practical collected data was conducted with the half⁃off cross test method. The experimental results show that the hit rate of the Top-N products in the association rule set is significantly higher than that of the interest recommendation method and the best-selling recommendation method.