Spaces:
Sleeping
Sleeping
Updated README.md for sentiment analysis of multi-domain reviews
Browse files
README.md
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
---
|
2 |
-
title: Sentiment Analysis
|
3 |
emoji: π»
|
4 |
colorFrom: gray
|
5 |
colorTo: yellow
|
@@ -10,7 +10,6 @@ pinned: false
|
|
10 |
license: apache-2.0
|
11 |
short_description: Sentiment Analysis of reviews using Hugging Face Transformer
|
12 |
---
|
13 |
-
|
14 |
# π Multi-Domain Sentiment Analysis
|
15 |
|
16 |
This project is a **Multi-Domain Sentiment Analysis Application** built using [Gradio](https://www.gradio.app/) and [Hugging Face Transformers](https://huggingface.co/transformers/).
|
@@ -24,7 +23,6 @@ It allows users to analyze the sentiment of reviews across different domains, su
|
|
24 |
The app predicts whether the sentiment is **Positive** or **Negative** and shows the confidence score.
|
25 |
|
26 |
---
|
27 |
-
|
28 |
## π Features
|
29 |
- π **Multi-domain support** (General, Products, Movies, Food/Restaurant).
|
30 |
- π€ **Pretrained models from Hugging Face Hub** for accurate sentiment classification.
|
@@ -56,15 +54,18 @@ Each domain uses a domain-specific model from Hugging Face Hub:
|
|
56 |
- **Predicted Sentiment** (Positive/Negative)
|
57 |
- **Confidence Score**
|
58 |
|
59 |
-
|
60 |
|
|
|
61 |
Domain: Food/Restaurant
|
62 |
Sentiment: Negative
|
63 |
Confidence: 99.95%
|
|
|
64 |
|
65 |
|
66 |
---
|
67 |
|
|
|
68 |
## π οΈ Installation (Optional: For local testing)
|
69 |
|
70 |
If you want to run this project locally:
|
@@ -80,16 +81,15 @@ pip install -r requirements.txt
|
|
80 |
# Run the app
|
81 |
python app.py
|
82 |
|
83 |
-
|
84 |
-
---
|
85 |
|
86 |
## π¦ Requirements
|
87 |
|
88 |
The main dependencies are:
|
89 |
|
90 |
-
-
|
91 |
-
-
|
92 |
-
-
|
93 |
|
94 |
All dependencies are listed in **`requirements.txt`**.
|
95 |
|
@@ -97,13 +97,14 @@ All dependencies are listed in **`requirements.txt`**.
|
|
97 |
|
98 |
## π Acknowledgements
|
99 |
|
100 |
-
- π€
|
101 |
-
- π¨
|
102 |
-
|
103 |
-
|
104 |
-
|
105 |
-
|
106 |
-
|
|
|
107 |
|
108 |
|
109 |
|
|
|
1 |
---
|
2 |
+
title: Multi-Domain Sentiment Analysis
|
3 |
emoji: π»
|
4 |
colorFrom: gray
|
5 |
colorTo: yellow
|
|
|
10 |
license: apache-2.0
|
11 |
short_description: Sentiment Analysis of reviews using Hugging Face Transformer
|
12 |
---
|
|
|
13 |
# π Multi-Domain Sentiment Analysis
|
14 |
|
15 |
This project is a **Multi-Domain Sentiment Analysis Application** built using [Gradio](https://www.gradio.app/) and [Hugging Face Transformers](https://huggingface.co/transformers/).
|
|
|
23 |
The app predicts whether the sentiment is **Positive** or **Negative** and shows the confidence score.
|
24 |
|
25 |
---
|
|
|
26 |
## π Features
|
27 |
- π **Multi-domain support** (General, Products, Movies, Food/Restaurant).
|
28 |
- π€ **Pretrained models from Hugging Face Hub** for accurate sentiment classification.
|
|
|
54 |
- **Predicted Sentiment** (Positive/Negative)
|
55 |
- **Confidence Score**
|
56 |
|
57 |
+
**Example:**
|
58 |
|
59 |
+
```
|
60 |
Domain: Food/Restaurant
|
61 |
Sentiment: Negative
|
62 |
Confidence: 99.95%
|
63 |
+
```
|
64 |
|
65 |
|
66 |
---
|
67 |
|
68 |
+
|
69 |
## π οΈ Installation (Optional: For local testing)
|
70 |
|
71 |
If you want to run this project locally:
|
|
|
81 |
# Run the app
|
82 |
python app.py
|
83 |
|
84 |
+
```
|
|
|
85 |
|
86 |
## π¦ Requirements
|
87 |
|
88 |
The main dependencies are:
|
89 |
|
90 |
+
- [Transformers](https://huggingface.co/docs/transformers/index)
|
91 |
+
- [PyTorch](https://pytorch.org/)
|
92 |
+
- [Gradio](https://www.gradio.app/)
|
93 |
|
94 |
All dependencies are listed in **`requirements.txt`**.
|
95 |
|
|
|
97 |
|
98 |
## π Acknowledgements
|
99 |
|
100 |
+
- π€ [Hugging Face](https://huggingface.co/) β for models and hosting.
|
101 |
+
- π¨ [Gradio](https://www.gradio.app/) β for the simple and elegant UI.
|
102 |
+
|
103 |
+
**Pretrained models used:**
|
104 |
+
- [DistilBERT](https://huggingface.co/distilbert-base-uncased)
|
105 |
+
- [NLPTown BERT](https://huggingface.co/nlptown/bert-base-multilingual-uncased-sentiment)
|
106 |
+
- [IMDB DistilBERT](https://huggingface.co/textattack/distilbert-base-uncased-imdb)
|
107 |
+
- [RoBERTa (Siebert)](https://huggingface.co/siebert/sentiment-roberta-large-english)
|
108 |
|
109 |
|
110 |
|