时空数据挖掘中的模式探讨

Pattern Discussion on Spatio Temporal Data Mining

随着数据的采集、存储、计算等能力不断提升,在数据挖掘领域,基于大数据的时空信息数据的分析和处理正是当前研究的热门问题。步入"互联网+"时代后,从巨大体量的时空大数据中挖掘出潜藏的有价值的信息具有重大意义。加之时空数据处理更为复杂,日趋繁重的时空数据处理任务急需寻找有效时空数据挖掘方法。据此,从数据挖掘中的时空数据挖掘模式的分析和展示角度出发,探讨时空数据挖掘的几种模式,包括时空频繁模式、时空关联模式、时空共现模式、时空分类、时空聚类、时空异常模式检测等,分析这些时空数据挖掘模式目前发展状况,对存在的问题及可能的解决办法进行探讨。

Nowadays, the ability of collecting, storing and computing data is rapidly improved. In the field of data mining, the analysis and processing of the spatio-temporal data based on large data is the hot research field. Into the ＂Internet ＋ ＂era, from the huge volume of spatio-temporal data, dig out the hidden valuable information has a great significance. In addition, the spatio-temporal data processing is more complex. Increasingly heavy spatio-temporal data processing tasks are in urgent need of finding effective spatio-temporal data mining methods. Focuses on the spatio temporal data mining analysis and display mode, several models of spatio-temporal data mining, including spatio temporal frequent patterns, spatiabtemporal correlation model, spatio temporal patterns of co-occurrence, spatio-temporal classification, spatio temporal clustering, spatial temporal outlier pattern detection is mainly discussed, the current development status of the spati^temporal data mining model is analyzed, and existing problems and possible solutions are discussed.