一种时序特征数据的分布式处理方法及系统

Distributed Processing Method and System for Time Series Feature Data

互联网用户数据中,其中一类是具有时序特征、记录用户操作过程的数据,隐含了用户的行为特征、偏好等信息。这类信息对互联网运营商改善服务品质、提升用户体验、优化运营策略有极大帮助。但是,这些信息无法直接得到,必须通过深入挖掘数据才能获取。基于此,阐述了一种时序特征数据的分布式处理方法及系统,高效挖掘此类数据中隐含的信息。

Among the Internet user data,one is the data with time series characteristics,which records the user's operation process,implying the user^s behavior characteristics,preferences and other information.This kind of information is very helpful for Internet operators to improve service quality,enhance user experience and optimize operation strategy.However,these information can not be obtained directly,and it must be obtained through in-depth data mining.Based on this,a distributed processing method and system for time series feature data is presented,which can efficiently mine the hidden information in such data.