Journal cover Journal topic
Atmospheric Measurement Techniques An interactive open-access journal of the European Geosciences Union
Journal topic

Journal metrics

IF value: 3.668
IF 5-year value: 3.707
IF 5-year
CiteScore value: 6.3
SNIP value: 1.383
IPP value: 3.75
SJR value: 1.525
Scimago H <br class='widget-line-break'>index value: 77
Scimago H
h5-index value: 49
Volume 10, issue 2
Atmos. Meas. Tech., 10, 695–708, 2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.
Atmos. Meas. Tech., 10, 695–708, 2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Research article 03 Mar 2017

Research article | 03 Mar 2017

Evaluation of machine learning algorithms for classification of primary biological aerosol using a new UV-LIF spectrometer

Simon Ruske1, David O. Topping1,2, Virginia E. Foot4, Paul H. Kaye5, Warren R. Stanley5, Ian Crawford1, Andrew P. Morse3, and Martin W. Gallagher1 Simon Ruske et al.
  • 1Centre for Atmospheric Science, SEAES, University of Manchester, Manchester, UK
  • 2NCAS, National Centre for Atmospheric Science, University of Manchester, Manchester, UK
  • 3Department of Geography and Planning, University of Liverpool, Liverpool, UK
  • 4Defence, Science and Technology Lab., Porton Down, Salisbury, Wiltshire, SP4 0JQ, UK
  • 5Particle Instruments Research Group, University of Hertfordshire, Hatfield, AL 10 9AB, UK

Abstract. Characterisation of bioaerosols has important implications within environment and public health sectors. Recent developments in ultraviolet light-induced fluorescence (UV-LIF) detectors such as the Wideband Integrated Bioaerosol Spectrometer (WIBS) and the newly introduced Multiparameter Bioaerosol Spectrometer (MBS) have allowed for the real-time collection of fluorescence, size and morphology measurements for the purpose of discriminating between bacteria, fungal spores and pollen.

This new generation of instruments has enabled ever larger data sets to be compiled with the aim of studying more complex environments. In real world data sets, particularly those from an urban environment, the population may be dominated by non-biological fluorescent interferents, bringing into question the accuracy of measurements of quantities such as concentrations. It is therefore imperative that we validate the performance of different algorithms which can be used for the task of classification.

For unsupervised learning we tested hierarchical agglomerative clustering with various different linkages. For supervised learning, 11 methods were tested, including decision trees, ensemble methods (random forests, gradient boosting and AdaBoost), two implementations for support vector machines (libsvm and liblinear) and Gaussian methods (Gaussian naïve Bayesian, quadratic and linear discriminant analysis, the k-nearest neighbours algorithm and artificial neural networks).

The methods were applied to two different data sets produced using the new MBS, which provides multichannel UV-LIF fluorescence signatures for single airborne biological particles. The first data set contained mixed PSLs and the second contained a variety of laboratory-generated aerosol.

Clustering in general performs slightly worse than the supervised learning methods, correctly classifying, at best, only 67. 6 and 91. 1 % for the two data sets respectively. For supervised learning the gradient boosting algorithm was found to be the most effective, on average correctly classifying 82. 8 and 98. 27 % of the testing data, respectively, across the two data sets.

A possible alternative to gradient boosting is neural networks. We do however note that this method requires much more user input than the other methods, and we suggest that further research should be conducted using this method, especially using parallelised hardware such as the GPU, which would allow for larger networks to be trained, which could possibly yield better results.

We also saw that some methods, such as clustering, failed to utilise the additional shape information provided by the instrument, whilst for others, such as the decision trees, ensemble methods and neural networks, improved performance could be attained with the inclusion of such information.

Publications Copernicus
Short summary
Particles such as bacteria, pollen and fungal spores have important implications within the environment and public health sectors. Here we evaluate the performance of various different methods for distinguishing between these different types of particles using a new instrument. We demonstrate that there may be better alternatives to the currently used methods which can be further investigated in future research.
Particles such as bacteria, pollen and fungal spores have important implications within the...