当前位置: X-MOL 学术Int. J. Syst. Assur. Eng. Manag. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Using cloud computing platform of 6G IoT in e-commerce personalized recommendation
International Journal of System Assurance Engineering and Management Pub Date : 2021-01-30 , DOI: 10.1007/s13198-021-01059-1
Junhai Wang , Yiman Zhang

In order to improve the exposure of commodities in e-commerce, accurately recommend personalized commodities and stimulate users' consumption of commodities, on the basis of a lot of literature, a personalized recommendation framework for e-commerce is built based on the open-source Hadoop cloud computing platform. According to the similarity of the algorithm, through the comparison of the project collaborative filtering algorithm based on cloud computing, user collaborative algorithm and the improved algorithm based on matrix filling and time context, the optimal algorithm is obtained and evaluated comprehensively in two aspects of algorithm performance and personalized recommendation performance. The results show that compared with UBCF (User Based Collaborative Filtering) and IBCF (Item Based Collaborative Filtering), the IA (Improved algorithm) based on matrix filling and time context is more accurate, its MAE (Mean absolute error) is smaller, and the scalability is enhanced. With the increase of nodes, the operation efficiency of the algorithm is obviously improved. The data recommendation effect of IA algorithm is better than that of IBCF and UBCF, and the commodities recommended to consumers are more in line with their preferences, which can improve the novelty and coverage rate of the algorithm, improve the exposure of the commodities in all aspects, and greatly promote the consumer behavior of users. The data mining system based on Hadoop cloud computing is introduced into e-commerce, which can provide theoretical basis for related research, and it is also of great significance to social development and academic research.



中文翻译:

在电子商务个性化推荐中使用6G IoT云计算平台

为了提高商品在电子商务中的曝光度,准确地推荐个性化商品并刺激用户对商品的消费,在大量文献的基础上,建立了基于开源的个性化电子商务推荐框架。 Hadoop云计算平台。根据算法的相似性,通过对基于云计算的项目协同过滤算法,用户协同算法和基于矩阵填充和时间上下文的改进算法的比较,从算法的两个方面综合获得了最优算法并进行了评估。性能和个性化推荐性能。结果表明,与UBCF(基于用户的协同过滤)和IBCF(基于项目的协同过滤)相比,基于矩阵填充和时间上下文的IA(改进算法)更准确,MAE(平均绝对误差)更小,可扩展性得到增强。随着节点的增加,该算法的运算效率明显提高。IA算法的数据推荐效果优于IBCF和UBCF,推荐给消费者的商品更符合他们的偏好,可以提高算法的新颖性和覆盖率,提高商品在各个领域的曝光率方面,大大促进了用户的消费行为。将基于Hadoop云计算的数据挖掘系统引入电子商务,可以为相关研究提供理论依据,对社会发展和学术研究也具有重要意义。

更新日期:2021-01-31
down
wechat
bug