Update README.md

README.md

@@ -9,10 +9,10 @@ app_file: app.py
 pinned: false
 license: apache-2.0
 ---
-# Sentiment Analysis with Voice Recommendations
+# SAFP: Sentiment Analysis Feedback Platform
 
 ## Project Objectives
-The goal of this project is to create an interactive web application that:
+The goal of SAFP (Sentiment Analysis Feedback Platform) is to create an interactive web application that:
 - Analyzes the sentiment of user-provided text, classifying it into a 1-5 star rating.
 - Provides tailored recommendations based on the detected sentiment.
 - Generates audio versions of the recommendations for an enhanced user experience.
@@ -22,7 +22,7 @@ The goal of this project is to create an interactive web application that:
 This app is designed as a demo to showcase sentiment analysis capabilities, making it accessible via Hugging Face Spaces for easy sharing and testing.
 
 ## Description of Implemented Pipelines
-The application consists of two main pipelines integrated into a Gradio interface:
+SAFP consists of two main pipelines integrated into a Gradio interface:
 
 1. **Sentiment Analysis Pipeline**:
    - **Technology**: Uses the `transformers` library from Hugging Face with the `pipeline` utility.
@@ -38,8 +38,8 @@ The application consists of two main pipelines integrated into a Gradio interfac
 These pipelines are combined in a single function (`analyze_sentiment`), which processes the input, generates the recommendation, and returns all outputs for display.
 
 ## Instructions for Using the Interface
-1. **Access the App**:
-   - Visit the Hugging Face Space URL (e.g., `https://huggingface.co/spaces/<your-username>/<space-name>` once deployed).
+1. **Access SAFP**:
+   - Visit the Hugging Face Space URL (e.g., `https://huggingface.co/spaces/<your-username>/SAFP` once deployed).
 
 2. **Input Your Comment**:
    - In the "Your Comment" textbox, type a short review or opinion (e.g., "The product is amazing!" or "الخدمة سيئة جداً").
@@ -57,13 +57,13 @@ These pipelines are combined in a single function (`analyze_sentiment`), which p
    - **Audio Recommendation**: Play the audio version of the recommendation.
 
 6. **Try Examples**:
-   - Use the provided examples at the bottom (e.g., "The product is amazing!" or "منتج جيد نوعاً ما") to test the app quickly.
+   - Use the provided examples at the bottom (e.g., "The product is amazing!" or "منتج جيد نوعاً ما") to test SAFP quickly.
 
-**Note**: The app may take a few seconds to load the model on first use.
+**Note**: SAFP may take a few seconds to load the model on first use.
 
 ## Justifications for Model and Pipeline Choices
 - **Model: `nlptown/bert-base-multilingual-uncased-sentiment`**:
-  - **Why Chosen**: This BERT-based model is pre-trained and fine-tuned for sentiment analysis on a 1-5 star scale, aligning perfectly with our goal of rating user comments. Its multilingual support handles both Arabic and English effectively.
+  - **Why Chosen**: This BERT-based model is pre-trained and fine-tuned for sentiment analysis on a 1-5 star scale, aligning perfectly with SAFP’s goal of rating user comments. Its multilingual support handles both Arabic and English effectively.
   - **Advantages**: High accuracy, robust handling of diverse text inputs, and availability on Hugging Face’s model hub for easy integration.
   - **Alternative Considered**: A simpler model (e.g., VADER) was an option but lacks multilingual support and nuanced star ratings.
 
@@ -72,7 +72,7 @@ These pipelines are combined in a single function (`analyze_sentiment`), which p
   - **Advantages**: Pre-built tokenization and prediction steps, making it ideal for quick deployment.
 
 - **Text-to-Speech Pipeline (`gTTS`)**:
-  - **Why Chosen**: Lightweight, free, and supports Arabic and English, matching our bilingual requirements. It integrates seamlessly with Gradio’s audio output.
+  - **Why Chosen**: Lightweight, free, and supports Arabic and English, matching SAFP’s bilingual requirements. It integrates seamlessly with Gradio’s audio output.
   - **Advantages**: No need for complex setup or API keys (unlike alternatives like AWS Polly), making it suitable for a demo app.
   - **Trade-off**: Audio quality is basic, but sufficient for this use case.
 
@@ -81,7 +81,7 @@ These pipelines are combined in a single function (`analyze_sentiment`), which p
   - **Advantages**: Free hosting, quick setup, and a professional look with minimal effort.
 
 ## Bilingual Implementation
-The app fully supports bilingual functionality for Arabic and English:
+SAFP fully supports bilingual functionality for Arabic and English:
 - **Sentiment Analysis**: The `nlptown/bert-base-multilingual-uncased-sentiment` model is multilingual, trained on diverse datasets, and can process text in either language without additional configuration. It detects sentiment regardless of the input language.
 - **Recommendations**: 
   - Predefined recommendations are stored in a dictionary with separate entries for Arabic and English.
@@ -91,16 +91,14 @@ The app fully supports bilingual functionality for Arabic and English:
   - This ensures the audio output aligns with the selected language.
 - **Interface**: The UI uses the `Tajawal` font (via CSS), which supports Arabic script and fallback to `Arial` for English, ensuring readability in both languages.
 
-**Justification**: Bilingual support was a core requirement to make the app accessible to Arabic and English speakers. The chosen tools naturally support this without needing separate models or complex logic, keeping the implementation efficient and maintainable.
+**Justification**: Bilingual support was a core requirement to make SAFP accessible to Arabic and English speakers. The chosen tools naturally support this without needing separate models or complex logic, keeping the implementation efficient and maintainable.
 
 ---
 
 ## Setup for Hugging Face Spaces
-To deploy this app:
-1. Upload `app.py` (the main script) and `requirements.txt` (dependencies: `gradio`, `transformers`, `gtts`, `torch`) to a Hugging Face Space.
-2. Set the hardware to "CPU Basic" (free tier, 2 vCPUs, 16 GB RAM), sufficient for this app’s lightweight needs.
-3. The Space will build and host the app, providing a public URL.
+To deploy SAFP:
+1. Upload `app.py` (the main script) and `requirements.txt` (dependencies: `gradio`, `transformers`, `gtts`, `torch`) to a Hugging Face Space named "SAFP".
+2. Set the hardware to "CPU Basic" (free tier, 2 vCPUs, 16 GB RAM), sufficient for SAFP’s lightweight needs.
+3. The Space will build and host SAFP, providing a public URL.
 
-For more details, see the deployment logs or contact the repository owner.
-
----
+For more details, see the deployment logs or contact the repository owner.