摘要
【目的/意义】大数据环境下,如何对海量的移动图书馆数据资源进行挖掘、重组和深度融合,从而获取最有价值的信息对移动图书馆的长远发展至关重要。【方法/过程】构建了大数据深度融合的移动图书馆情境化推荐系统,通过深度融合图书馆用户的情境信息,有效缓解大数据环境下评分数据稀疏导致的推荐性能下降问题;同时采用MapReduce的并行处理方式,以此提高大数据的融合与挖掘性能。【结果/结论】实验结果表明,大数据深度融合的移动图书馆情境化推荐系统较好地将情境信息融入到移动图书馆知识推荐过程中,改进了推荐性能,有利于为用户提供精准的个性化服务资源,MapReduce化的并行处理方法也有效提升了大数据融合与挖掘的性能与效率。
【Purpose/significance】Under the big data environment,how to excavate,reorganize and integrate the massive mobile library data resources to obtain the most valuable information is very important to the long-term development of mobile library.【Method/process】In this paper,a mobile library situational recommendation system with large data deep fusion is constructed,which can effectively alleviate the recommendation performance decline caused by sparse data sparsity in large data environment by deep fusion of library users’situation information.We also use the parallel processing mode of MapReduce to improve the performance of the fusion and mining of large data.【Result/conclusion】The experimental results show that the mobile library situational recommendation system can integrate the situation information into the knowledge recommendation process and improve the recommendation performance.The MapReduce parallel processing method also effectively improves the performance and efficiency of the large data fusion and mining.
作者
刘海鸥
黄文娜
苏妍嫄
张亚明
LIU Hai-ou;HUANG Wen-na;SU Yan-yuan;ZHANG Ya-ming(Internet+and Industrial Development Research Center.Yanshan University,Qinhuangdao 066004,China;School of Economics and Management,Yanshan University,Qinhuangdao 066004,china)
出处
《情报科学》
CSSCI
北大核心
2019年第1期68-73,共6页
Information Science
基金
国家社科基金项目“基于大数据深度融合的移动图书馆用户画像情境化推荐模型研究”(18BTQ033)
关键词
大数据
移动图书馆
情境化推荐
big data
mobile digital library
situational recommendation
