Environmental Quality Short-term Prediction

• Purpose
• Correlation Pattern Prediction in Time and Space
• Data loader
• Applications
• Configuration
• Operating environment
• Precautions
• Reference

Purpose

This information asset collects various environmental observation data, learns and predicts air quality (AQI) ranks based on their spatiotemporal correlation patterns, and generates map data (risk maps) ^[1] from the prediction results.The process includes the collection of various environmental observation data.

Correlation Pattern Prediction in Time and Space

AQI rank prediction is based on the CRNN deep learning model ^[2].CRNN first learns the spatial correlation patterns between observed data and AQI ranks using Convolution Neural Networks (CNN).Then, Long Short Term Memory (LSTM), a type of Recurrent Neural Networks (RNN), is used to learn temporal correlation patterns.Long Short Term Memory (LSTM), a type of Recurrent Neural Networks (RNN), is used to learn temporal correlation patterns. This CRNN machine learning model is used to predict the AQI rank in the target area one to several hours later. The predicted results are used to generate a risk map that can be displayed on a map.

Data loader

Environmental observation data used for short-term environmental quality prediction is collected from the Internet and other sources, and data for analysis in the form of event data is extracted and registered in a database.

Applications

• Alert notifications based on short-term environmental quality forecasts.
• Safe route guidance, etc., that can help mitigate traffic risks due to extreme weather conditions.

Configuration

This information asset consists of.

• Processing program（Python）
  • CRNN Predictive Model Generation Program
  • CRNN predictive model training program
  • CRNN Predictive Model Evaluation Program
  • CRNN Predictive Model Testing Program
  • CRNN Predictive Model Transfer Learning Program
  • AQI Short-term Forecasting Program
• Data model
  • Measurement station data table (public format)
  • Healthcare sensor data table (event format)
  • Atmospheric data table (public format)
  • CRNN prediction model information table (public format)
  • CRNN prediction model generation table (public format)
  • CRNN prediction model evaluation table (public format)
  • CRNN predictive model test table (public format)
  • CRNN Predictive Model Transfer Learning Tables (public format)
  • AQI short-term prediction table (public format)
• Predictive model (CRNN^[2] method)
  • AQI (PM2.5) short-term forecast model (5th order regional mesh, hourly)

Operating environment

We confirmed the operation in the following environment.

• OS : Ubuntu 18.04 LTS
• RAM : 4GB Recommended
• DISK SPACE : 100GB or more
• Internet connection available
• Python : 3.x（Does not work with 2.x）
  • Python library
  • psycopg2
  • configparser
  • geojson
  • requests
• PostgreSQL : 9.2 or more

When using the xDataEdge^[3]environment, the following conditions are required in addition to the above.

• Services such as Apache, Nginx, etc. that have HTTP server functions (LISTEN to HTTP/HTTPS ports) must not be running on the OS.
• Have root privileges on the OS where xData Edge is installed

Precautions

• Restrictions
  • Functional Limitations
    • This information asset is not guaranteed to operate in environments other than those shown in the operating environment.
  • Data Restrictions
    • There are restrictions on the data to be registered in the database. For details, please refer to the Data Loader API Manual^[4].
• Disclaimer
  • While every effort has been made to ensure that the information on this page is as accurate as possible, we do not guarantee its accuracy or safety.
  • We are not responsible for any damages caused by the contents of this page or by what is provided.
  • All information provided on this page is sample only and we are not responsible for any predicted results.

Reference

• [1] DCCS Sample Applications:https://dccs-trial.nicttb-b5g.jp/dccs_sample_aqi/
• [2] Zhao, P. and Zettsu, K.: Convolution Recurrent Neural Networks for Short-Term Prediction of Atmospheric Sensing Data, The 4th IEEE International Conference on Smart Data (SmartData 2018), Halifax, Canada, pp.815-821 (July 2018).
• [3] xData Edge manual : https://www.xdata.nict.jp/docs/Edge/1.0/.
• [4] Data Loader API manual : https://www.xdata.nict.jp/docs/DataLoader/0.6/.