Abstract: In order to anticipate and classify bioassay data labeled AID362, this study investigates the use of sophisticated machine learning techniques, including neural networks and extra trees. The usefulness and suitability of these algorithms for bioassay categorization and regression are examined by means of meticulous testing and analysis. The UCI repository dataset is prepared by applying several data preprocessing techniques such as cleaning, normalization, and feature scaling. Overfitting is reduced and dimensionality is made easier with the help of Extra Trees feature extraction. The effectiveness of the trained models in managing high-dimensional, nonlinear bioassay data is demonstrated by a variety of performance indicators. The findings add significant knowledge to the field by demonstrating the effectiveness of neural networks in identifying complex patterns and the advantages of extra trees in managing complicated datasets.

Keywords: Bioassay, AID362, Classification, Regression.

| DOI: 10.17148/IARJSET.2024.11624