Update README.md

README.md

@@ -4,7 +4,7 @@ emoji: 💬
 colorFrom: yellow
 colorTo: purple
 sdk: gradio
-sdk_version: 5.42.0
+sdk_version: 6.2.0
 app_file: app.py
 pinned: false
 hf_oauth: true
@@ -14,4 +14,57 @@ license: mit
 short_description: 'Interactive demo of phishing email detection with AI, based '
 ---
 
-An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
+## 🧠 How it works
+
+- The user provides the raw text of an email.
+- The text is processed by a **fine-tuned BERT model** for binary classification:
+  - `PHISHING`
+  - `SAFE`
+- The model outputs a label and a confidence score.
+- A threshold-based policy is applied:
+  - High-confidence phishing → 🚨 **PHISHING**
+  - High-confidence safe → 🟢 **LOW RISK**
+  - Intermediate confidence → 🟠 **REVIEW recommended**
+
+This approach reflects a **security-oriented mindset**, where uncertain cases are intentionally flagged for manual review.
+
+---
+
+## 🔬 Model and data
+
+- **Model**: `ElSlay/BERT-Phishing-Email-Model`
+- **Task**: Text classification (phishing vs safe)
+- **Dataset**:  
+  [`zefang-liu/phishing-email-dataset`](https://huggingface.co/datasets/zefang-liu/phishing-email-dataset)
+
+The model is used in **inference-only mode**; no training is performed within this demo.
+
+---
+
+## 🎯 Project context
+
+This demo is part of a broader experimental effort related to **PhishForge**, an evolving framework focused on phishing analysis and threat detection using AI-driven techniques.
+
+The goal is to explore how NLP models can be integrated into practical cybersecurity workflows in a transparent and interpretable way.
+
+---
+
+## 📌 Notes and limitations
+
+- The analysis is **content-based only**.
+- Email headers, metadata, URLs reputation, and attachments are not evaluated.
+- The demo is intended for **educational and experimental purposes**, not for production use.
+
+---
+
+## 📄 Citation
+
+A citable version of this project is available via Zenodo: https://huggingface.co/spaces/giulcs008/phishing-email-detector
+
+
+---
+
+## 👤 Author
+
+**Giulia Casaldi**  
+Cybersecurity & AI