Table 2 Multi-class classification report for logistics regression (LR), decision trees (DT) and support vector machines (SVM) tested on the computational battery compounds data set, encoded using one-hot, Mendeleev and Mendeleev+ encoding
From: Classification of battery compounds using structure-free Mendeleev encodings
Algorithm | Encoding | Metric | Class 0 | Class 1 | Class 2 | Class 3 | Class 4 |
|---|---|---|---|---|---|---|---|
| Â | Â | Precision | 0.971 | 0.987 | 1.000 | 0.974 | 0.991 |
| Â | One-hot | Recall | 0.973 | 0.984 | 0.992 | 1.000 | 0.983 |
| Â | Â | Accuracy | 0.972 | 0.986 | 0.996 | 0.987 | 0.987 |
| Â | Â | Precision | 0.989 | 1.000 | 0.955 | 0.950 | 0.975 |
LR | Mendeleev | Recall | 0.985 | 0.980 | 0.980 | 1.000 | 0.965 |
| Â | Â | Accuracy | 0.987 | 0.990 | 0.968 | 0.974 | 0.970 |
| Â | Â | Precision | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Mendeleev+ | Recall | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Â | Accuracy | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Â | Precision | 0.990 | 0.995 | 1.000 | 0.982 | 0.982 |
| Â | One-hot | Recall | 0.983 | 0.995 | 1.000 | 1.000 | 0.991 |
| Â | Â | Accuracy | 0.987 | 0.995 | 1.000 | 0.991 | 0.987 |
| Â | Â | Precision | 0.998 | 1.000 | 0.997 | 1.000 | 0.990 |
DT | Mendeleev | Recall | 1.000 | 0.983 | 1.000 | 1.000 | 0.997 |
| Â | Â | Accuracy | 0.999 | 1.000 | 0.990 | 1.000 | 0.993 |
| Â | Â | Precision | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Mendeleev+ | Recall | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Â | Accuracy | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Â | Precision | 0.993 | 0.997 | 1.000 | 0.991 | 0.983 |
| Â | One-hot | Recall | 0.993 | 0.996 | 1.000 | 1.000 | 0.991 |
| Â | Â | Accuracy | 0.993 | 0.997 | 1.000 | 0.996 | 0.987 |
| Â | Â | Precision | 0.998 | 0.990 | 0.994 | 1.000 | 0.991 |
SVM | Mendeleev | Recall | 0.998 | 1.000 | 0.986 | 1.000 | 0.995 |
| Â | Â | Accuracy | 0.998 | 0.995 | 0.990 | 1.000 | 0.993 |
| Â | Â | Precision | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Mendeleev+ | Recall | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |
| Â | Â | Accuracy | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 |