Skip to main content
. 2022 May 25;12(11):1810. doi: 10.3390/nano12111810

Figure 7.

Figure 7

Graphical network of the 19 articles that meet the criteria to be included in Table 1. (a) Network with all the connections between the types of nanomaterials, their relevant structural characteristics (descriptors), and the QSAR models used to predict a defined endpoint. (b) Most relevant connections of metal oxide class. (c) Most relevant connections of optimal quasi-SMILES-based descriptors and the closely related Monte Carlo (MC) algorithm. ANN: artificial neural network; DecTrees: decision trees; GBR: gradient boosting regressor; KNN: k-nearest neighbors; LinReg: linear regression; LogReg: logistic regression; MC: Monte Carlo; MLR: multiple linear regression; PLS: partial least squares; RF: random forest; SVM: support vector machine; XGBoost: extreme gradient boosting.