摘要
现阶段大多是利用社交网络理论进行分析,发现话单数据中的潜在人员。社交网络理论是将数据中的实体用节点表示,节点间的关系用线表示。大数据时代很多传统的算法在针对多特征的数据时,分析结果的理想性越来越差。而机器学习这几年在数据分析工作中大放异彩,为很多经典问题提供了一种新的解决思路。论文正是基于这样的背景,提出了一种新的推荐算法,用图嵌入的方法研究话单数据,将通话网络中的点和关系向量化,使机器学习算法用于话单分析成为可能。
The potential people are found in the data of call list is mostly using the theory of social networking. The social network theory is to represent the entities in the data as nodes,and the relationships between the nodes are represented by lines. In the era of big data,many traditional algorithms are more and more poor in analyzing data with multiple features. In recent years,machine learning has made great achievements in data analysis,providing a new solution to many classical problems. Based on this background,this paper proposes a new recommendation algorithm. It uses graph embedding method to study bill data,vectorizes the points and relationships in the call network,and makes it possible to use machine learning algorithms for analysis call list.
作者
韩文轻
彭艳兵
HAN Wenqing;PENG Yanbing(Nanjing FiberHome World Communication Technology Co.,Ltd.,Nanjing 210019;Wuhan Research Institute of Posts and Telecommunications,Wuhan 430074)
出处
《计算机与数字工程》
2020年第2期393-397,共5页
Computer & Digital Engineering
关键词
话单分析
机器学习
图嵌入
call list’s analysis
machine learning
graph embedding
作者简介
韩文轻,女,硕士研究生,研究方向:数据挖掘;彭艳兵,男,博士,硕士生导师,研究方向:网络行为分析、海量数据挖掘。
