Cloud infrastructures of JINR and its Member States’ organizations/ Data analysis – Machine Learning in air pollution research

Seminars

Laboratory of Information Technologies

Joint Laboratory Seminar

Date and Time: Wednesday, 2 February 2022, at 3:00 PM

Venue: Online seminar on Webex, Laboratory of Information Technologies

  1. Seminar topic: «Cloud infrastructures of JINR and its Member States’ organizations»

    Speaker: N. Kutovskiy

    Abstract:

    The current state of affairs on the JINR cloud infrastructure and distributed platform based on the resources of the organizations of the Institute’s Member States is presented. A description of the architectures of both platforms, technical characteristics, tasks for which their resources are used, as well as development plans, are provided.

  2. Seminar topic: «Data analysis – Machine Learning in air pollution research»

    Speaker: V. Svozilík

    Abstract:

    Vast area air pollution research is intensive on financial, human and computational resources. Therefore, the trend to find alternative approaches for studying air pollution can be observed.
    Land Use Regression (LUR) is an alternative modelling methodology based on the assumption that concentration at a particular location is determined by factors (land use, pollution sources) that positively or negatively influence air pollution. The relation between these factors and air pollution concentration can be represented via a regression model. The regression model was constructed using an Artificial Neural Network (ANN), which is able to capture nonlinear phenomena. The results showed that ANN-assisted LUR described more variability of air pollution than standard LUR models.
    The second experiment focuses on the improvement of air pollution monitoring data from a low-cost sensor network. Based on the measurement of low-cost sensors carried out with a standard air pollution monitoring station simultaneously, the ANN is able to improve measurements of the whole sensor network.