Data preprocessing was performed to improve dataset quality by removing duplicate entries, descriptors with missing values, near-zero variance, and high pairwise correlation (r > 0.8) to reduce multic

Data preprocessing was performed to improve dataset quality by removing duplicate entries, descriptors with missing values, near-zero variance, and high pairwise correlation (r > 0.8) to reduce multicollinearity. Feature selection was then applied to reduce dimensionality and enhance model performance while minimising overfitting. can u find me refernce for r value