Skip to main content
Fig. 4 | BMC Medical Informatics and Decision Making

Fig. 4

From: Developing a novel causal inference algorithm for personalized biomedical causal graph learning using meta machine learning

Fig. 4

Overall results: The X axis denotes the size of the graph d, and the Y axes are the three types of evaluation metrics, FDR (false discovery rate), SHD (structural Hamming distance), and NNZ (number of non-zero entries). When we fix the number of samples per patient/task and the graph type, we can see that MetaNoTears-L1 outperforms others in SHD consistently. Although MTL has a similar performance with MetaNoTears-L1 on FDR, it has a higher SHD and also predicts a much lower number of edges. In addition, as the size of the graph increases, MetaNoTears-L1’s performance worsens slowest

Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us