Sentiment Analysis Using Pre-Trained BERT
Problem Statement
Develop a Transformer-based sentiment analysis model using a pre-trained DistilBERT model to classify IMDb movie reviews as Positive or Negative. The model will use the Hugging Face Transformers library for tokenization and fine-tuning and will be evaluated using suitable classification metrics.
Dataset Description
The IMDb Movie Review Dataset is a widely used dataset for binary sentiment classification. Each record contains a movie review and its corresponding sentiment label.
| Attribute | Description |
|---|---|
| Dataset | IMDb Movie Reviews |
| Data Type | Text |
| Task | Sentiment Analysis / Text Classification |
| Input | Movie review text |
| Target | Sentiment label |
| Label 0 | Negative Review |
| Label 1 | Positive Review |
| Training Set | 25,000 labeled reviews |
| Test Set | 25,000 labeled reviews |
| Number of Classes | 2 |
In the Hugging Face version, the two important fields are
text and label, where 0 represents negative and 1 represents positive. 
0 टिप्पण्या
कृपया तुमच्या प्रियजनांना लेख शेअर करा आणि तुमचा अभिप्राय जरूर नोंदवा. 🙏 🙏