![]() ![]() GI-Forum 1, 65–81 (2018)īudde, M., et al.: Smartaqnet: remote and in-situ sensing of urban air quality. Keywordsīruns, J., Riesterer, J., Wang, B., Riedel, T., Beigl, M.: Automated quality assessment of (citizen) weather stations. In contrast to existing approaches, we can model the uncertainty of the prediction based on the quality and quantity of input signals crawled. Furthermore, we also validate the contribution of ultra-low-cost sensors in the SmartAQnet sensor network, which reduces the average MAE of our model pipeline from 4.18 \(\upmu \)g/m \(^\) and increases the PCC from 0.589 to 0.665. And predicts the daily average PM10 mass concentration for the next day. Our approach is based on a neural kernel network deep kernel learning model that takes the highly heterogeneous and uncertain measurements provided by a post-hoc hybrid low-cost sensor network as input. In this paper, we propose a scheme to deal with measurement sources of different quality for time-series prediction of urban particulate matter. Modern web technologies allow novel types of sensor networks that collect measurements from different sources ranging from citizen-collected data to official sources.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |