Bioagriculture Outlier Elimination Based on 3D View of X - Y Variance and Leverage Measurement

Abstract EN

Aiming at effective outlier elimination in the biological near-infrared spectral andachieving high accuracy predictive modeling, this paper proposes a novel outlier eliminationmethod based on X - Y variance and leverage analysis.

Firstly, the characters of near-infraredspectral are summarized; then residual sample X -variance, leverage, and residual sample Y -variance are concatenated as a divergence measurement.

We further compared theproposed method with X - Y variance, Mahalanobis distance, and Hotelling T 2 statistical analysis; the experiment results demonstrate that the proposed methods have competitive outlier elimination and better performance in time complexity and accuracy.

The proposed methodcan also be adopted for other outlier elimination tasks.