Automatic mass spectra recognition for Ultra High Vacuum systems using multilabel classification

作者：

Highlights：

• We propose an application of multilabel methods to mass-spectrum identification.

• The best performing model was a dependency binary classifier using XGBoost.

• The mean and lowest AUC for the test set were 0.986 and 0.89, respectively.

• A public interactive web has been created with the best model’s functionalities.

摘要

•We propose an application of multilabel methods to mass-spectrum identification.•The best performing model was a dependency binary classifier using XGBoost.•The mean and lowest AUC for the test set were 0.986 and 0.89, respectively.•A public interactive web has been created with the best model’s functionalities.

论文关键词：Residual gas analysis,Mass spectrum identification,Multilabel classification,Ultra High Vacuum,Outgassing

论文评审过程：Received 29 July 2020, Revised 21 December 2020, Accepted 25 March 2021, Available online 1 April 2021, Version of Record 18 April 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114959