• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of information and computational science >Research on Individual Influence in Social Networking Services Based on MapReduce
【24h】

Research on Individual Influence in Social Networking Services Based on MapReduce

机译：基于MapReduce的社交网络服务中个人影响力研究

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

With the development of Social Networking Services (SNS), mining and analyzing data from SNS is becoming an active area of science. We propose UAAPeR (Unequal-Authorities-Assignment-PeopleRank) algorithm in this paper, based on the PageRank algorithm and the limitations of its unequal authorities assignment. This method can handle user group data of large-scale efficiently, and rank individual influence for a certain group quickly and accurately. At the same time, the algorithm can be applied to the framework of MapReduce of big data processing platform, for example, Hadoop. In each iterative process of the algorithm, relational file of user-focus can be analyzed by Map function. Score of individual influence can be computed by Reduce function. Thereby intermediate iterative process of PageRank algorithm parallelization can be conducted. Accurate ranking results can be obtained from mass data in the end. The experimental results show that the UAAPeR algorithm is more accurate than PageRank algorithm, also with better cluster scalability and faster execution speed.
机译：随着社交网络服务（SNS）的发展，对来自SNS的数据进行挖掘和分析正成为科学的活跃领域。基于PageRank算法及其不平等的权限分配的局限性，本文提出了UAAPeR（Unequal-Authorities-Assignment-PeopleRank）算法。该方法可以有效地处理大规模的用户群数据，并快速，准确地对特定群体的个人影响力进行排名。同时，该算法可以应用于Hadoop等大数据处理平台的MapReduce框架。在算法的每个迭代过程中，可以通过Map函数分析用户关注的关系文件。个人影响力的分数可以通过减少功能来计算。从而可以进行PageRank算法并行化的中间迭代过程。最后，可以从大量数据中获得准确的排名结果。实验结果表明，UAAPeR算法比PageRank算法更准确，并且具有更好的集群可扩展性和更快的执行速度。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号