Update README.md

README.md

@@ -16,6 +16,11 @@ license: mit
 
 Unlike traditional filters (e.g., "Italian", "$$"), VEN uses **Semantic Search** and **Persona Matching** to understand the *intent* behind the search (e.g., "A romantic quiet place for a date" vs. "Lively place for drinks with friends").
 
+## 📸 Interface Preview
+
+![Main Interface](WhatsApp%20Image%202026-01-19%20at%2022.28.58.jpeg)
+*(The VEN interface features a clean dark-mode design with smart filtering)*
+
 ## 🚀 Key Features
 
 * **Semantic Search:** Uses `Sentence-Transformers` (NLP) to convert user preferences into vector embeddings.
@@ -32,12 +37,11 @@ Unlike traditional filters (e.g., "Italian", "$$"), VEN uses **Semantic Search**
 * **Scikit-Learn:** For calculating Cosine Similarity between user input and reviewer profiles.
 * **Pandas & NumPy:** For data manipulation and vector operations.
 
-## 📂 Project Structure
+## 🖼️ More Screenshots
 
-* `app.py`: The main application logic (UI, Search Engine, Logic).
-* `cleaned_dataset_10k.csv`: A processed dataset of 10,000 restaurant reviews from Tel Aviv.
-* `final_embeddings_10k.pkl`: Pre-computed vector embeddings for the reviews (saves processing time).
-* `requirements.txt`: List of dependencies required to run the app.
+| Recommendation Card | Filter Settings | Search Results |
+|:-------------------:|:---------------:|:--------------:|
+| ![Card](WhatsApp%20Image%202026-01-19%20at%2022.28.46.jpeg) | ![Filters](WhatsApp%20Image%202026-01-19%20at%2022.28.34%20(1).jpeg) | ![Results](WhatsApp%20Image%202026-01-19%20at%2022.28.22%20(1).jpeg) |
 
 ## 🧠 How It Works