Table 5 Resulting average classification accuracy together with precision, recall and F1-score for each MLP model tested in our approach to performing the symbolic anchoring functionalities in nuScene, MOTFront and Mix datasets and in the Leon@Home scenario.
From: SAILOR: perceptual anchoring for robotic cognitive architectures
Accuracy | Precision | Recall | F1-score | |
|---|---|---|---|---|
nuScenes dataset | ||||
nuScenes | 0.9918 | 1.0000 | 0.7574 | 0.8619 |
MOTFront | 0.8697 | 0.9987 | 0.5849 | 0.7377 |
Mix | 0.9077 | 0.9988 | 0.5930 | 0.7441 |
Leon@Home | 0.9627 | 0.9923 | 0.6400 | 0.7781 |
MOTFront dataset | ||||
nuScenes | 0.9873 | 1.0000 | 0.6260 | 0.7700 |
MOTFront | 0.9982 | 0.9976 | 0.9966 | 0.9971 |
Mix | 0.9948 | 0.9976 | 0.9793 | 0.9884 |
Leon@Home | 0.9914 | 0.9953 | 0.9200 | 0.9562 |
Mix dataset | ||||
nuScenes | 0.9882 | 0.8776 | 0.7574 | 0.8131 |
MOTFront | 0.9966 | 0.9916 | 0.9977 | 0.9946 |
Mix | 0.9940 | 0.9870 | 0.9865 | 0.9867 |
Leon@Home | 0.9927 | 0.9832 | 0.9453 | 0.9639 |
