基于大数据的用户特征分析

摘要：
互联网应用到各行各业，用户在使用各种不同业务的同时产生并积累了大量的历史数据。海量用户数据中蕴藏着丰富的信息，已经成为计算机时代最宝贵的资源。应运而生的数据挖掘技术和云计算技术旨在挖掘用户大数据中蕴含的价值。用户数据中包含用户行为特征，而用户行为通常和多种社会因素和技术参数相关，这些会影响不同场景下用户的角色及特征规律。各学科中，对一组对象的研究都是基于实测时间序列，并通过各种数学手段对其进行处理，寻找出序列变化特征、发展规律与趋势，从而对未来某时刻的状态进行估计。本文对运营商用户数据进行分析，从现有用户数据所具备的特征出发，深入挖掘用户行为规律，并通过构建时间序列分析模型，寻找用户特征相关因子，梳理出影响用户行为的单因素及多因素，分析用户特征规律，获取用户的行为模式，从而创造更多的价值。

关键词：
大数据；时间序列；用户特征

WU Guanfang^1,， CUI Hongyan^2,*

（

1、Information and Communication Engineering School,Beijing University of Posts and Telecommunications,State Key Lab. of Networking and Switching Technology,Key Lab. of Network System Architecture Convergence,Beijing,100876；       2、2.Information and Communication Engineering School,Beijing University of Posts and Telecommunications,State Key Lab. of Networking and Switching Technology,Key Lab. of Network System Architecture Convergence,Beijing,100876；  ）

Abstract：
It has accumulated a large number of user data in the process of the infiltration from internet to industry. Vast amount of user data contains a wealth of information, which has become the most valuable resources of the computer era. Data mining technology and cloud computing technology are designed to tap value for user data. User data contains user behavior characteristics, and user behavior is usually associated with a variety of social factors and technical parameters, which will affect the user's role and characteristics of different scenarios.In each subject, the research on a group of objects is measured based on time series, and carries on the processing through mathematics, to find out the law of development and trend of sequence variation, and thus to estimate the future state of a moment. This paper carries on the analysis to the user data of the operators, starting with features from the existing user data, dig the user behavior, and find the factors related to user characteristics by constructing time series model, sort out the single factor and multiple factors affecting consumer behavior. By analysing user characteristics, it can obtain user behavior patterns, thereby creating more value.

Keywords：
big data; time series; user characteristics