Advertisement

A Width-Variable Window Attention Model for Environmental Sensors

  • Cuiqin Hou
  • Yingju Xia
  • Jun Sun
  • Jing Shang
  • Ryozo Takasu
  • Masao Kondo
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10635)

Abstract

Air pollution is a major problem in modern cities and developing countries. Fine particulate matter (PM2.5) is a growing public health concern and become the most serious air pollution. In this study, we formulate the PM2.5 inference problem in conventional environmental sensors as a sequence-to-sequence problem. We adopt the encoder-decoder LSTM (Long short term memory) framework to solve the PM2.5 inference problem. A novel width-variable window attention mechanism is proposed for the encoder-decoder LSTM system. The proposed method learn the position and width of the attention window simultaneously. The proposed method is evaluated on large scale data and the experimental results show that it achieves better performance on two datasets with different concentration of PM2.5.

Keywords

LSTM Attention model RNN 

References

  1. 1.
    Dockery, D.W., Xu, X., Spengler, J.D.: An association between air pollution and mortality in six U.S. cities. N. Engl. J. Med. 329, 1753–1759 (1993)CrossRefGoogle Scholar
  2. 2.
    Pope, C.A., Bates, D.V., Raizenne, M.E.: Health effects of particulate air pollution: time for reassessment. Environ. Health Perspect. 103, 472–480 (1995)CrossRefGoogle Scholar
  3. 3.
    Mage, D., Ozolins, G., Peterson, P.: Urban air pollution in megacities of the world. Atmos. Environ. 30, 681–686 (1996)CrossRefGoogle Scholar
  4. 4.
    Hamra, G.B., Guha, N., Cohen, A.: Outdoor particulate matter exposure and lung cancer: a systematic review and meta-analysis. Environ. Health Perspect. 122, 906–912 (2014)Google Scholar
  5. 5.
    Takasu, R.: Development of compact and low-cost PM2.5 mass concentration measurement equipment and its application to high-frequency mobile monitoring in a local area. In: 2016 International Conference on Agriculture, Energy and Environment Engineering, Bangkok (2016)Google Scholar
  6. 6.
    Oprea, M., Mihalache, S.F., Popescu, M.: A comparative study of computational intelligence techniques applied to PM2. 5 air pollution forecasting. In 6th International Conference on Computers Communications and Control, Oradea, Romania, pp. 103–108 (2016)Google Scholar
  7. 7.
    Park, D., Kwon, S.B., Cho, Y.: Development and calibration of a particulate matter measurement device with wireless sensor network function. Int. J. Environ. Monit. Anal. 1, 15–20 (2013)CrossRefGoogle Scholar
  8. 8.
    Gao, M., Cao, J., Seto, E.: A distributed network of low-cost continuous reading sensors to measure spatiotemporal variations of PM2. 5 in Xi’an, China. Environ. Pollut. 199, 56–65 (2015)CrossRefGoogle Scholar
  9. 9.
    Dong, Y., Wang, H., Zhang, L.: An improved model for PM2. 5 inference based on support vector machine. In: 17th International Conference on Software Engineering, Artificial Intelligence, Shanghai, China, pp. 27–31 (2016)Google Scholar
  10. 10.
    Srimuruganandam, B., Nagendra, S.M.S.: ANN-based PM prediction model for assessing the temporal variability of PM10, PM2. 5 and PM1 concentrations at an urban roadway. Int. J. Environ. Eng. 7, 60–89 (2015)CrossRefGoogle Scholar
  11. 11.
    Zheng, Y., Liu, F., Hsieh, H.P.: U-Air: When urban air quality inference meets big data. In: 19th ACM SIGKDD International Conference on Knowledge discovery and Data Mining, New York, pp. 1436–1444 (2013)Google Scholar
  12. 12.
    Coffey, C.C., Pearce, T.A.: Direct-reading methods for workplace air monitoring. J. Chem. Health Saf. 17, 10–21 (2010)CrossRefGoogle Scholar
  13. 13.
    Bahdanau, D., Cho, K.H., Bengio, Y.: Neural Machine Translation by Jointly Learning to Align and Translate. arXiv preprint arXiv:1409.0473 (2014)
  14. 14.
    Chorowski, J.K., Bahdanau, D., Serdyuk, D.: Attention-based models for speech recognition. In: 29th Annual Conference on Neural Information Processing Systems, Quebec, Canada, pp. 577–585 (2015)Google Scholar
  15. 15.
    Gregor, K., Danihelka, I., Graves, A.: DRAW: A recurrent neural network for image generation. arXiv preprint arXiv:1502.04623 (2015)

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Cuiqin Hou
    • 1
  • Yingju Xia
    • 1
  • Jun Sun
    • 1
  • Jing Shang
    • 2
  • Ryozo Takasu
    • 3
  • Masao Kondo
    • 1
  1. 1.Fujitsu Research and Development Center Co., LTD.BeijingChina
  2. 2.College of Environmental Sciences and EngineeringPeking UniversityBeijingChina
  3. 3.Fujitsu Laboratories LTD.KawasakiJapan

Personalised recommendations