Evolutionary optimisation with outlier detection-based deep learning model for biomedical data classification

R. Raja and B. Ashok

International Journal of Networking and Virtual Organisations, 2022, vol. 27, issue 2, 143-162

Abstract: In recent times, large amount of medical data is being generated by various sources such as test reports, medications, etc. Due to the recent advances of machine learning (ML) and deep learning (DL) models, medical data classification (MDC) remains a crucial process in the healthcare sector. This study introduces a new hyperparameter tuned convolutional neural network-recurrent neural network (HPT-CNN-RNN) model for medical data classification. The proposed HPT-CNN-RNN model includes pre-processing step to transform the actual healthcare data into useful format. Besides, SVM-SMOTE approach was executed to handle the class imbalance problems. In addition, outlier detection process is performed using extreme gradient boosting (XGBoost) model. Moreover, bacterial foraging optimisation algorithm (BFOA) with CNNRNN model is employed to categorise medical data. Furthermore, the BFOA is utilised to optimally choose the hyperparameter values of the CNNRNN model. The experimental outcomes designated the better performance of the HPT-CNN-RNN model over the other methods.

Keywords: classification; medical data; data mining; outlier detection; class imbalance; deep learning; parameter tuning. (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.inderscience.com/link.php?id=127606 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ids:ijnvor:v:27:y:2022:i:2:p:143-162

Access Statistics for this article

More articles in International Journal of Networking and Virtual Organisations from Inderscience Enterprises Ltd
Bibliographic data for series maintained by Sarah Parker ().