论文部分内容阅读
随着互联网电子商务的高速发展,推荐系统在电子商务领域得到了广泛的应用。煤炭产业也开始引进了电子销售系统。在煤炭系统中,推荐系统利用消费者对消费商品的排名打分,分析相似性并进一步预测消费者可能感兴趣的商品。协同过滤算法被普遍应用在推荐系统中。但是,煤炭销售数据规模逐渐增大,传统的协同过滤算法不能有效地处理海量规模煤炭数据,推荐效率很低。本文针对大规模煤炭销售数据,提出了基于Mapreduce的分布式协同过滤算法,该算法有效地完成推荐系统的预测及推荐工作。通过大量的实验结果也进一步表明本文提出的算法与传统算法相比,具有很高的效率,并且扩展性良好。
With the rapid development of Internet e-commerce, the recommendation system has been widely used in the field of e-commerce. The coal industry has also started to introduce electronic sales systems. In the coal system, the recommender system uses consumer rankings of consumer goods, analyzes similarities and further predicts what consumers may be interested in. Collaborative filtering algorithms are commonly used in recommendation systems. However, the scale of coal sales data is gradually increasing. The traditional collaborative filtering algorithm can not effectively deal with mass coal data, and the recommended efficiency is very low. In this paper, based on the large-scale coal sales data, a distributed cooperative filtering algorithm based on Mapreduce is proposed, which can effectively predict and recommend the recommended system. Through a large number of experimental results also further show that the proposed algorithm compared with the traditional algorithm, with high efficiency, and good scalability.