Advertisement

New Modification Version of Principal Component Analysis with Kinetic Correlation Matrix Using Kinetic Energy

  • Sara K. Al-Ruzaiqi
  • Christian W. Dawson
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 886)

Abstract

Principle Component Analysis (PCA) is a direct, non-parametric method for extracting pertinent information from confusing data sets. It presents a roadmap for how to reduce a complex data set to a lower dimension to disclose the hidden, simplified structures that often underlie it. However, most PCA methods are not able to realize the desired benefits when they handle real world, and nonlinear data. In this work, a modified version of PCA with kinetic correlation matrix using kinetic energy is proposed. The features of this modified PCA have been assessed on different data sets of air passenger numbers. The results show that the modified version of PCA is more effective in data compression, classes reparability and classification accuracy than using traditional PCA.

Keywords

Principle Component Analysis (PCA) Kinetic correlation matrix Kinetic energy Algorithm Prediction 

Notes

Acknowledgment

I would like to address my special acknowledgements to all those people who provide me with data for my experiments. My warm appreciation is due to the Public Authority for Civil Aviation, Directorate General of Meteorology, and Ministry of Tourism in Oman.

References

  1. 1.
    Bengio, Y.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1798–1828 (2013)CrossRefGoogle Scholar
  2. 2.
    Timmerman, M.E.: Principal Component Analysis, 2nd edn. Springer, New York (2003). I. T. Jolliffe. Journal of the American Statistical Association, 98, 1082-1083Google Scholar
  3. 3.
    Palechor, F.M., et al.: Cardiovascular disease analysis using supervised and unsupervised data mining techniques. J. Softw. 12(2), 81–90 (2017)Google Scholar
  4. 4.
    Van der Maaten, L., Hinton, G.: Visualizing data using t-SNE. J. Mach. Learn. Res. 9(2579–2605), 85 (2008)zbMATHGoogle Scholar
  5. 5.
    Coates, A., Ng, A.Y.: Learning feature representations with k-means. In: Neural Networks (2012)Google Scholar
  6. 6.
    Chen, T.Q., Guestrin, C.: XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016, pp. 785–794 (2016)Google Scholar
  7. 7.
    Lowe, D.G.: Object recognition from local scale-invariant features. Paper presented at the proceedings of the seventh ieee international conference on computer vision, 1999 (1999)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Computer Science DepartmentLoughborough UniversityMuscatOman
  2. 2.Computer Science DepartmentLoughborough UniversityLoughboroughUK

Personalised recommendations