Text Classification is the task of assigning a label or class to a given text. Some use cases are sentiment analysis, natural language inference, and assessing grammatical correctness.
Stanford University has obtained characteristics of billions of words that have strange proportions. Click on the link to read more and download.
Using NLP, I want to classify the sentences and display a special emoji based on the emotion that the sentence evokes.
- Dataset
](https://github.com/KhadijehValipour/Text_Classification/blob/main/assents/spaces_u_image.png)
- Network
pip install -r requirements.txt
python Emoji_Text_Classification.py --dimension 100 --sentence "I am very well today"
| Vectors | Accuracy | Loss | Accuracy with dropout | Loss with dropout | Inference average time |
|---|---|---|---|---|---|
| 50d | 0.81 | 0.6 | 0.80 | 0.72 | 0.11s |
| 100d | 0.90 | 0.4 | 0.82 | 0.61 | 0.13s |
| 200d | 0.94 | 0.2 | 0.82 | 0.48 | 0.12s |
| 300d | 0.99 | 0.1 | 0.96 | 0.41 | 0.12s |
