% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@ARTICLE{Stadtler:906258,
      author       = {Stadtler, Scarlet and Betancourt, Clara and Roscher,
                      Ribana},
      title        = {{E}xplainable {M}achine {L}earning {R}eveals
                      {C}apabilities, {R}edundancy, and {L}imitations of a
                      {G}eospatial {A}ir {Q}uality {B}enchmark {D}ataset},
      journal      = {Machine learning and knowledge extraction},
      volume       = {4},
      number       = {1},
      issn         = {2504-4990},
      address      = {Basel},
      publisher    = {MDPI},
      reportid     = {FZJ-2022-01329},
      pages        = {150 - 171},
      year         = {2022},
      abstract     = {Air quality is relevant to society because it poses
                      environmental risks to humans and nature. We use explainable
                      machine learning in air quality research by analyzing model
                      predictions in relation to the underlying training data. The
                      data originate from worldwide ozone observations, paired
                      with geospatial data. We use two different architectures: a
                      neural network and a random forest trained on various
                      geospatial data to predict multi-year averages of the air
                      pollutant ozone. To understand how both models function, we
                      explain how they represent the training data and derive
                      their predictions. By focusing on inaccurate predictions and
                      explaining why these predictions fail, we can (i) identify
                      underrepresented samples, (ii) flag unexpected inaccurate
                      predictions, and (iii) point to training samples irrelevant
                      for predictions on the test set. Based on the
                      underrepresented samples, we suggest where to build new
                      measurement stations. We also show which training samples do
                      not substantially contribute to the model performance. This
                      study demonstrates the application of explainable machine
                      learning beyond simply explaining the trained model.},
      cin          = {JSC},
      ddc          = {004},
      cid          = {I:(DE-Juel1)JSC-20090406},
      pnm          = {5111 - Domain-Specific Simulation $\&$ Data Life Cycle Labs
                      (SDLs) and Research Groups (POF4-511) / IntelliAQ -
                      Artificial Intelligence for Air Quality (787576) / AI
                      Strategy for Earth system data $(kiste_20200501)$ / Earth
                      System Data Exploration (ESDE)},
      pid          = {G:(DE-HGF)POF4-5111 / G:(EU-Grant)787576 /
                      $G:(DE-Juel1)kiste_20200501$ / G:(DE-Juel-1)ESDE},
      typ          = {PUB:(DE-HGF)16},
      UT           = {WOS:000774979600001},
      doi          = {10.3390/make4010008},
      url          = {https://juser.fz-juelich.de/record/906258},
}