metadata

language: ko
tags:
  - medical
  - bert
  - korean
license: apache-2.0
datasets:
  - other
model-index:
  - name: Medical Chatbot
    results:
      - task:
          name: Medical Text Classification
          type: text-classification
        metrics:
          - type: accuracy
            value: 0.95
            name: Validation Accuracy

Medical Chatbot Model

This model is a fine-tuned version of KLUE BERT base for medical symptom classification.

Model description

The model uses KLUE BERT base architecture fine-tuned on a custom medical dataset. It performs classification for various diseases and provides treatment recommendations.

Input Format

{
    "text": "두통이 심하고 메스꺼워요"
}

Output Format

{
    "label": "편두통",
    "score": 0.95
}

Training procedure

Epochs: 5
Batch size: 8
Learning rate: 2e-5
Maximum sequence length: 128

Limitations and bias

This model should be used for reference only and not as a substitute for professional medical advice.