Mining and Forecasting of Big Time-Series Data
Mining and Forecasting of Big Time-Series Data
复制标题
DOI：
10.1145/2723372.2731081
复制
发表时间：
2015-05
期刊：
影响因子：
--
通讯作者：
Yasushi Sakurai;Yasuko Matsubara;C. Faloutsos
中科院分区：
文献类型：
--
作者：
Yasushi Sakurai;Yasuko Matsubara;C. Faloutsos
给定大量的时间序列，例如运动捕捉传感器和汽车轨迹，我们如何高效地找到典型模式？我们如何统计总结所有序列，并实现有意义的分割？预测和异常值检测的主要工具是什么？由于硬件成本的降低和在线处理能力的提高，时间序列数据分析变得越来越重要。我们项目的目标是开发大时间序列数据实时建模和预测的基础技术。我们提供了这些强大技术背后的直觉，并介绍了说明其实际用途的案例研究。
Given a large collection of time series, such as motion capture sensors and automobile trajectories, how can we efficiently and effectively find typical patterns? How can we statistically summarize all the sequences, and achieve a meaningful segmentation? What are the major tools for fore-casting and outlier detection? Time-series data analysis becomes of increasingly high importance, thanks to the decreasing cost of hardware and the increasing online processing abilities. The objective of our project is to develop fundamental technologies for the real-time modeling and forecasting of big time-series data. We provide the intuition behind these powerful technologies, as well as to introduce case studies that illustrate their practical use.