Data Generation Using Gene Expression Generator

• Zakarya Farou ^[1] ; Noureddine Mouhoub ^[2] ; Tomáš Horváth ^{[1] [3]}
  1. [1] Eötvös Loránd University

     Eötvös Loránd University

     Hungría

     
  2. [2] University of Bordeaux

     University of Bordeaux

     Arrondissement de Bordeaux, Francia

     
  3. [3] University of Pavol Jozef Šafárik

     University of Pavol Jozef Šafárik

     Eslovaquia

     

  Mostrar afiliaciones +
• Localización: Intelligent Data Engineering and Automated Learning – IDEAL 2020. 21st International Conference: Guimarães, Portugal; November 4–6, 2020. Proceedings / Cesar Analide (ed. lit.), Paulo Novais (ed. lit.), David Camacho Fernández (ed. lit.), Hujun Yin (ed. lit.), Vol. 2, 2020 (Part II), ISBN 978-3-030-62365-4, págs. 54-65
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Generative adversarial networks (GANs) could be used efficiently for image and video generation when labeled training data is available in bulk. In general, building a good machine learning model requires a reasonable amount of labeled training data. However, there are areas such as the biomedical field where the creation of such a dataset is timeconsuming and requires expert knowledge. Thus, the aim is to use data augmentation techniques as an alternative to data collection to improve data classification. This paper presents the use of a modified version of a GAN called Gene Expression Generator (GEG) to augment the available data samples. The proposed approach was used to generate synthetic data for binary biomedical datasets to train existing supervised machine learning approaches. Experimental results show that the use of GEG for data augmentation with amodified version of leave one out cross-validation (LOOCV) increases the performance of classification accuracy.