Table 2 Comparison of the top-performing models for each approach in classifying medication state
Medication State Classification Approach | Model | Accuracy | Precision | Recall | F1 macro |
|---|---|---|---|---|---|
Semantic Search | multi-qa-mpnet-base-dot-v1 | 0.848 | 0.847 | 0.848 | 0.848 |
Machine Learning | gte-Qwen2-1.5B-instruct. + Random Forests, gte-Qwen2-1.5B-instruct. + Support Vector Machine | 0.924 | 0.925 | 0.924 | 0.924 |
LLMs | Gemma-2 (9B), Llama-3.1 (8B) (few-shot setting) | 0.984 | 0.984 | 0.984 | 0.984 |
