Advertisement

An Unsupervised Machine Learning Analysis of the FIRST Radio Sources

  • David Bastien
  • Radhakhrishna Somanah
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 863)

Abstract

The large availability of radio sources from the Faint Images of the Radio Sky at Twenty cm (FIRST) has inspired us to use unsupervised Machine Learning (ML) to do a morphological segmentation of 1000 radio sources. Through techniques like shapelets decomposition, we were able to decompose each radio sources into a series of 256 coefficients that were input into unsupervised ML techniques like Isometric Mapping (ISOMAP) for dimensionality reduction and density-based spatial clustering of applications with noise (DBSCAN) as clustering algorithm. Through this process we were able to identify four groups of sources and 189 outliers. After comparing the segmentation results with our human classification, we found that the method achieved an accuracy of 0.83, with an \(F_1\) score of 0.87. Showing that unsupervised ML could be used to classify images in the radio astronomy domain.

Keywords

Unsupervised machine learning Isomap DBSCAN Radio astronomy Shapelets analysis 

References

  1. 1.
    Ball, N.M., Brunner, R.J.: Data mining and machine learning in astronomy. Int. J. Mod. Phys. D 19, 1049–1106 (2010)CrossRefGoogle Scholar
  2. 2.
    Bastien, D., Oozeer, N., Somanah, R.: Classifying bent radio galaxies from a mixture of point-like/extended images with machine learning. IOP Conf. Ser.: Mater. Sci. Eng. 198, 012013 (2017)Google Scholar
  3. 3.
    Becker, R.H., White, R.L., Helfand, D.J.: The FIRST survey: faint images of the radio sky at twenty centimeters. APJ, 450, 559 (1995)Google Scholar
  4. 4.
    Dayan, P.: Unsupervised Learning. The MIT Encyclopedia of Cognitive Sciences. The MIT Press, London (1999)Google Scholar
  5. 5.
    Ester, M., Kriegel, H., Sander, J., Xu, X.: A density-based algorithm for discovering clusters. In: KDD-96 Proceedings, pp. 226–231 (1996)Google Scholar
  6. 6.
    Gautam, A., et al.: An improved mammogram classification approach using back propagation neural network. Data Engineering and Intelligent Computing. Springer, Singapore, pp. 369–376 (2018)Google Scholar
  7. 7.
    Helfand, D.J., White, R.L., Becker, R.H.: The last of FIRST: the final catalog and source identifications. Astrophys. J. 801, 26 (2015)CrossRefGoogle Scholar
  8. 8.
    Lay-Ekuakille, A., et al.: Multidimensional analysis of EEG features using advanced spectral estimates for diagnosis accuracy. In: 2013 IEEE International Symposium on Medical Measurements and Applications Proceedings (MeMeA). IEEE (2013)Google Scholar
  9. 9.
    Le, D.N., et al.: Optimizing feature selection in video-based recognition using MaxMin Ant System for the online video contextual advertisement user-oriented system. J. Comput. Sci. 21, 361–370 (2017)CrossRefGoogle Scholar
  10. 10.
    Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., et al.: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)MathSciNetzbMATHGoogle Scholar
  11. 11.
    Proctor, D.D.: Comparing pattern recognition feature sets for sorting triples in the FIRST database. APJS 165, 5–107 (2006)Google Scholar
  12. 12.
    Proctor, D.D.: Morphological annotations for groups in the first database. Astrophys. J. Suppl. Ser. 194, 31 (2011)CrossRefGoogle Scholar
  13. 13.
    Refregier, A.: Shapelets - I. A method for image analysis. MNRAS 338, 35–47 (2003)CrossRefGoogle Scholar
  14. 14.
    Samudrala, S.K., Zola, J., Aluru, S., Ganapathysubramanian, B.: Parallel framework for dimensionality reduction of large-scale datasets. Sci. Program. 2015, 1–12 (2015)Google Scholar
  15. 15.
    Tenenbaum, J.B., de Silva, V., Langford, J.C.: A global geometric framework for nonlinear dimensionality reduction. Science 5500, 35–47 (2003)Google Scholar
  16. 16.
    Tiwari, A., et al.: ANN-based classification of mammograms using nonlinear preprocessing. In: Proceedings of 2nd International Conference on Micro-Electronics, Electromagnetics and Telecommunications. Springer, Singapore (2018)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.Hydrus Labs LtdRoches Brunes, Rose HillMauritius
  2. 2.Universite des MascareignesRoche Brunes, Rose HillMauritius
  3. 3.University of MauritiusReduitMauritius

Personalised recommendations