Figure 5
From: Functional random forest with applications in dose-response predictions
Performance comparison for Functional Random Forest and Random Forest for both synthetic data and CCLE data. (a) For noisy synthetic data, FRF can follow the actual response variations even though it was modeled using noisy data while RF fails to follow the trend in higher dose levels, (b) For fitted dose-response curve in CCLE Liver cell line SNU449 post Erlotinib administration, FRF prediction again outperforms RF prediction.
