Internet of Things Technology and LSTM-based Data Imputation and Prediction for Transportation Industry

Abstract

The Corona Virus Disease 2019 (COVID-19) has touched individuals from all walks of life in recent years, and the movement of people and commodities has become a crucial route for the disease's spread. Reasonable control of the transportation industry has become a vital to the epidemic's prevention and control. Simultaneously, a succession of tight blockade inspection measures in the transportation business have significantly restricted the movement of people and commodities, putting a strain on the industry. As a result, it is crucial to study the transportation industry's comprehensive measurement in this environment for future regulation and prevention. In this paper, the transportation industry's freight and passenger transport are used as indicators to assess the economy, and the data source is multi-source time series data from government statistics. To anticipate future economic trends in the transportation business, we chose freight and passenger transportation in the air, land, and marine industries, respectively. Due to the significant amount of missing data, this paper develops a time series data imputation approach for its specific missing situations in order to fill in the gaps and make future prediction tasks easier. In addition, this paper builds a long and short-term memory network to train the data in order to predict future transportation industry economic trends.