A Bibliometric Survey on the Use of Long Short-Term Memory Networks for Multivariate Time series forecasting

Authors

Vidur Sood Mr., Symbiosis Institute of Technology, Symbiosis International (Deemed University)Follow
Manobhav Mehta Mr., Symbiosis Institute of Technology, Symbiosis International (Deemed University)Follow
Vedansh Mishra Mr., Symbiosis Institute of Technology, Symbiosis International (Deemed University)Follow
Akash Upadhyay Mr., Symbiosis Institute of Technology, Symbiosis International (Deemed University)Follow
Shilpa Hudnurkar, Symbiosis Institute of Technology, Symbiosis International (Deemed University)Follow
Shilpa Gite Dr., Symbiosis Institute of Technology, Symbiosis International (Deemed University)Follow
Neela Rayavarapu Dr., Symbiosis International (Deemed University)Follow

Date of this Version

5-20-2021

Abstract

In this paper, we aim to review and analyze the publications related to the utilization of Long Short-Term Memory (LSTM) networks for multivariate time series forecasting. The purpose of this bibliometric survey was to study how technology in the field of LSTM has evolved over the years. There were 242 research papers published, by over 50 researchers, over 6 years, on the topic of “Multivariate time series forecasting using LSTM”. The majority of these papers were published between the years 2018 and 2020. The Scopus database was utilized for analyzing recent trends in this area and to determine the model that would be best suited for weather forecasting applications. Through this study, we aim to shortlist various models that have shown consistent reliability and accuracy while utilizing multivariate time series data for prediction. These models can then be employed for other forecasting applications.