metadata

tags:
  - ernie
  - health
  - tweet
datasets:
  - custom-phm-tweets
metrics:
  - accuracy
model-index:
  - name: ernie-phmtweets-sutd
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: custom-phm-tweets
          type: labelled
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.885

ernie-phmtweets-sutd

This model is a fine-tuned version of ernie-2.0-en for text classification to identify public health events through tweets. The project was based on an Emory University Study on Detection of Personal Health Mentions in Social Media paper, that worked with this custom dataset.

It achieves the following results on the evaluation set:

Accuracy: 0.885

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("dibsondivya/ernie-phmtweets-sutd")
model = AutoModelForSequenceClassification.from_pretrained("dibsondivya/ernie-phmtweets-sutd")

Model Evaluation Results

With Validation Set

Accuracy: 0.889763779527559

With Test Set

Accuracy: 0.884643644379133

References for ERNIE 2.0 Model

@article{sun2019ernie20,
  title={ERNIE 2.0: A Continual Pre-training Framework for Language Understanding},
  author={Sun, Yu and Wang, Shuohuan and Li, Yukun and Feng, Shikun and Tian, Hao and Wu, Hua and Wang, Haifeng},
  journal={arXiv preprint arXiv:1907.12412},
  year={2019} 
}