Research Article
Dynamic Time Warping Based Clustering for Time Series Analysis
@INPROCEEDINGS{10.1007/978-3-030-94182-6_29, author={Kun Zhang and Shuai Lin and Haoxuan Sun and Liyao Ma and Junpeng Xu}, title={Dynamic Time Warping Based Clustering for Time Series Analysis}, proceedings={IoT and Big Data Technologies for Health Care. Second EAI International Conference, IoTCare 2021, Virtual Event, October 18-19, 2021, Proceedings, Part II}, proceedings_a={IOTCARE PART 2}, year={2022}, month={6}, keywords={K-MEANS DTW Similarity Dynamic weighting}, doi={10.1007/978-3-030-94182-6_29} }- Kun Zhang
Shuai Lin
Haoxuan Sun
Liyao Ma
Junpeng Xu
Year: 2022
Dynamic Time Warping Based Clustering for Time Series Analysis
IOTCARE PART 2
Springer
DOI: 10.1007/978-3-030-94182-6_29
Abstract
This paper proposes a prediction method based on time series similarity. Based on clustering, Dynamic Time Warping (DTW) algorithm is used to find the influence of similarity and weight on the prediction results. Time series is a structure that records data in time sequence. The characteristics of multiple data at each time point are the same and comparable. According to people’s purpose to find the rule of time series, and to the future time forecast. The first chapter introduces the background of the topic. The second chapter mainly introduces the time series, clustering algorithm, similarity, DTW distance and other basic theories involved in this paper. In the third chapter, we study the method of the total forecast data of time series. DTW distance is used for clustering to obtain the similarity with each class and then predict the data.
