1(College of Information Science and Technology,Nanjing 210095,China）2(Nanjing Agricultural University National Engineering and Technology Center for Infomation Agriculture,Nanjing 210095,China）3(Jiangsu Collaborative Center for the Technology and Application of Internet of Things,Nanjing 210023,China)
Abstract：With the development of machine learning,new technologies continue to emerge in the field of timeseries. How to efficiently analyze the internal patterns of time series and extract the identifiable characteristics of time series is becoming a research hotspot.This article first introduces the latest developments in TS research,then analyze and compare the research situation of machine learning methods on time series in detail from the aspects of waveform extraction,timedependent characteristics,and sequence transformation of the feature extraction algorithm,and finally based on the development trend of the current time series feature extraction algorithm,the future development of the time series feature extraction algorithm is prospected.
2021, Vol. 42 
