摘要
针对基于经典协同过滤算法的点餐推荐系统中的数据稀疏性问题,通过加入Apriori关联规则算法并融合基于内容的相似度,进行菜品评分预测,填充评分矩阵,降低数据的稀疏度;并结合点餐个性化需求,设置基于人数的推荐标准,进一步过滤推荐列表;经与User-CF、Item-CF的对比实验,改进后的系统有效地解决了经典协同过滤算法中的数据稀疏性问题,推荐效果更好和很好的泛化性能。
In order to solve the problem of data sparsity in order-to-order recommendation system based on classical collaborative filtering algorithm,Apriori Association rule algorithm is added and the similarity degree based on content is fused to predict the food score,fill the score matrix and reduce the sparsity of the data;Combined with the personalized needs of meal ordering,the recommendation standard based on the number of people is set to further filter the recommendation list.After the comparison experiment with User-CF and Item-CF,the improved system effectively solves the problem of data sparsity in the classical collaborative filtering algorithm,and has better recommendation effect and good generalization performance.
作者
饶刘维
叶强胜
代世佳
陈兴文
Rao Liuwei;Ye Qiangsheng;Dai Shijia;Chen Xingwen(School of Information and Communication Engineering,Dalian Minzu University,Dalian Liaoning 116600,China)
出处
《山西电子技术》
2024年第5期84-85,102,共3页
Shanxi Electronic Technology
关键词
推荐系统
Apriori关联规则
人数推荐
recommendation system
Apriori association rules
recommended number of people
