Efficient Query Processing in Time Series
Yuhong Li · 2015
With the rapid development over the last decade, time series data become one of the most frequently used data in real world applications (e.g., finance analysis, medical diagnosis, environmental monitoring, etc.). As expected, the volume of the time series data will even grow larger in near future. It is important to design efficient and effective algorithm and index to handle various tasks for these data. Thereby, my PhD study focuses on how to extract meaningful time series patterns from large volume of data efficiently. Specifically, two types of extraction queries are discussed in this work, including longest-lasting correlated subsequence query and time series motif query. The applications and solutions of these two queries are thoroughly introduced and discussed in this paper. Moreover, some potential pattern extraction queries will also be discussed in this paper.