融合朴素贝叶斯和协同过滤的外卖推荐并行算法研究

PARALLEL ALGORITHM FOR TAKEAWAY RECOMMENDATION WITH NAIVE BAYES AND COLLABORATIVE FILTERING

为了提高个性化外卖推荐系统的准确率,结合传统的协同过滤算法中存在的数据稀疏性问题,提出一种融合朴素贝叶斯和协同过滤的外卖推荐并行算法.采用并行的朴素贝叶斯分类算法构建外卖评论文本情感分类器,量化评论文本情感值;结合评分数值构建外卖综合评分模型;将综合评分结果整合到推荐系统的训练集,利用优化的并行ALS算法进行推荐.实验结果表明,该推荐算法不仅在推荐准确率上有一定的提高,还具有良好的加速比.该算法应用于个性化外卖推荐是可行和有效的.

In order to improve the accuracy of the personalized takeaway recommendation system,combined with the problem of data sparsity in the traditional collaborative filtering algorithm,this paper proposes a parallel algorithm for takeaway recommendation with naive Bayes and collaborative filtering.The parallel naive Bayes classification algorithm was used to construct a takeaway comment text sentiment classifier,which aimed to quantify the emotional value of the comment text.The comprehensive scoring model for takeaway was constructed by combining the scoring values.The comprehensive scoring results were integrated into the training set of the recommendation system,and the optimized parallel ALS algorithm was used for recommendations.Experimental results show that our algorithm not only has a certain improvement in the recommendation accuracy rate,but also has a good acceleration ratio.It is feasible and effective to apply the algorithm to personalized takeaway recommendation.