A novel probabilistic source apportionment approach: Bayesian auto-correlated matrix factorization

Rusanen, Anton; Björklund, Anton; Manousakas, Manousos I.; Jiang, Jianhui; Kulmala, Markku T.; Puolamäki, Kai; Daellenbach, Kaspar R.

doi:https://doi.org/10.5194/amt-17-1251-2024

Articles | Volume 17, issue 4

https://doi.org/10.5194/amt-17-1251-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-17-1251-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 17, issue 4

Research article

|

22 Feb 2024

Research article |

| 22 Feb 2024

A novel probabilistic source apportionment approach: Bayesian auto-correlated matrix factorization

Anton Rusanen, Anton Björklund, Manousos I. Manousakas, Jianhui Jiang, Markku T. Kulmala, Kai Puolamäki, and Kaspar R. Daellenbach

Data sets

Bayesian autocorrelated matrix factorization, datasets A. Rusanen, A. Björklund, M. I. Manousakas, J. Jiang, M. T. Kulmala, K. Puolamäki, and K. R. Daellenbach https://doi.org/10.5281/zenodo.10629577

Model code and software

Bayesian autocorrelated matrix factorization, software A. Rusanen, A. Björklund, M. I. Manousakas, J. Jiang, M. T. Kulmala, K. Puolamäki, and K. R. Daellenbach https://doi.org/10.5281/zenodo.10629849

Short summary

We present a Bayesian non-negative matrix factorization model that performs better on our test datasets than currently widely used models. Its advantages are better use of time information and providing a direct error estimation. We believe this could lead to better estimates of emission sources from measurements.