Performance of Time-Varying Particle Swarm Optimizer to Predict Cancers

  • T. R. Vijaya LakshmiEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 768)


Classification of tumors is a challenging task in the field of bioinformatics. The gene expression levels measured using microarray approach contains thousands of levels. Finding optimum number of genes expression levels to classify tumor samples is carried out in this paper using PSO. The conventional PSO algorithm works with constant social and cognitive coefficients. This paper proposes time-varying PSO in which the social and cognitive coefficients are allowed to vary with respect to time. The performance of the proposed particle swarm optimizer gives better results when compared to the conventional PSO in classifying the tumor samples.


Particle swarm optimization Time-varying coefficients Tumor classification Gene expression levels 


  1. 1.
    Gunavathi, C., Premalatha, K.: A comparative analysis of swarm intelligence techniques for feature selection in cancer classification. Sci. World J. 2014, Article ID 693831, (2014)
  2. 2.
    Alonso, C.G.J., Moro-Sancho, I.Q., Simon-Hurtado, A., Varela-Arrabal, R.: Microarray gene expression classification with few genes: Criteria to combine attribute selection and classification methods. Expert Syst. Appl. 39, 7270–7280 (2012)CrossRefGoogle Scholar
  3. 3.
  4. 4.
    Vijaya Lakshmi, T.R., Sastry, P.N., Rajinikanth, T.V.: Feature optimization to recognize Telugu handwritten characters by implementing DE and PSO techniques. In: International conference on FICTA, Springer, Odisha, India, pp. 397–405 (2016)Google Scholar
  5. 5.
    Vijaya Lakshmi, T.R., Sastry, P.N., Rajinikanth, T.V.: Feature selection to recognize text from palm leaf manuscripts. Signal, Image and Video Processing. Springer, Article in press, Berlin (2017). Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.MGITHyderabadIndia

Personalised recommendations