A SURVEY ON TIME SERIES PREDICTION USING MACHINE LEARNING

Abstract

This paper offers an summary of present day literature related to time series classification by
implementing using Deep Neural Network (DNN) related methods in initial time series for classifying
based on distinct distance procedures similar to Euclidean or active distances based on time warping as
the paper initiates by reviewing standard approaches of time series classification which aspires for
classifying time series through minimal chronological classifications that are potential for possessing
the minimal classification accuracy. Additionally most of the papers emphasis in influencing the
essential machine learning procedures as the process explores various machine learning tools that are
required to implement various projects of machine learning that describes chronological information
combined for predicting each model that highlights recent advancements in hybrid deep learning
models that are based on statistical models using with neural network constituents for improving each
of the categories of time series data.