Predicting antiviral compounds for avian influenza A/H9N2 using logistic regression with RBF Kernel

Amiroch, Siti, Jamhuri, Mohammad, Irawan, Mohammad Isa and Mukhlash, Imam Predicting antiviral compounds for avian influenza A/H9N2 using logistic regression with RBF Kernel. Presented at International Conference on Modeling & E-Information Research, Artificial Learning and Digital Applications (ICMERALDA).

Text 20434.pdf - Published Version Download (230kB)

Abstract

Avian Influenza A/H9N2 is a significant threat to the global poultry industry and presents occasional but severe health risks to humans. Given the potential ramifications of an outbreak, the swift and accurate identification of effective antiviral compounds becomes crucial. Traditional methods employed for predicting the efficacy of these compounds often encounter challenges, particularly in maintaining a balance between accuracy and efficiency. Recognizing these limitations, our study introduces an innovative predictive approach. We leverage the combined strengths of Radial Basis Function (RBF) networks and Logistic Regression. This methodology transforms compound features using the RBF network. The changed features are then fed into a Logistic Regression model to make predictions regarding efficacy. Initial findings from our research indicate a remarkable enhancement in prediction accuracy and precision compared to prevalent methods. Furthermore, our study provides a potentially transformative tool for antiviral compound prediction and establishes a precedent, emphasizing the profound potential of hybrid modeling techniques in advancing biomedical research.

Item Type:	Conference (Paper)
Keywords:	Avian Influenza A/H9N2; Hybrid machine learning models; Log-RBF methodology; Antiviral compound prediction; Drug repurposing
Subjects:	01 MATHEMATICAL SCIENCES > 0103 Numerical and Computational mathematics > 010302 Numerical Solution of Differential and Integral Equations
Divisions:	Faculty of Mathematics and Sciences > Department of Mathematics
Depositing User:	Mr Jamhuri Mohammad
Date Deposited:	01 Oct 2024 12:33

Downloads

Origin of downloads

Actions (login required)

View Item