摘要
随着互联网的快速发展,大量数据也随之产生,并且数据量级甚至达到了PB,ZB 级。那么如何在大数据环境中改进传统推荐算法已成为当务之急。为了一定程度上缓解传统推荐算法所遇到的稀疏的数据等问题。首先构建一种传递信任的模型,在此基础之上,提出一种基于信任的协同过滤推荐模型。最后,在大数据的背景下,针对海量数据时的可扩展性和计算效率问题,模型在Map Reduce处理平台上进行并行化处理,并进行了一系列的对比实验。实验结果证明,并行化处理后,模型在解决冷启动问题,减轻数据稀疏性和提高推荐精度的同时,可扩展性和计算效率也表现出良好的效果。
With the rapid development of Internet,a large number of data are produced,and the data level even reaches Pb,ZB level.So how to improve the traditional recommendation algorithm in big data environment has become an urgent task.In order to alleviate the problem of sparse data encountered by traditional recommendation algorithm to a certain extent.Firstly,a model of trust transfer is constructed,and then a collaborative filtering recommendation model based on trust is proposed.Finally,in the context of big data,aiming at the scalability and computing efficiency of massive data,the model is parallelized on the map reduce processing platform,and a series of comparative experiments are carried out.The experimental results show that after parallel processing,the model not only solves the cold start problem,reduces data sparsity and improves the recommendation accuracy,but also has good scalability and computing efficiency.
作者
李帅
李晓会
杜颖
Li Shuai;Li Xiaohui;Du Ying(College of electronics and information engineering,Liaoning University of Technology,Jinzhou Liaoning,121001)
出处
《电子测试》
2020年第11期72-75,共4页
Electronic Test
基金
国家自然科学基金青年基金项目(61802161)
辽宁省科技厅基金项目(20170540448)。
关键词
推荐算法
大数据
信任模型
可扩展性
并行化处理
recommendation algorithm
big data
trust model
scalability
parallel processing