Data Mining – Tunisian House Pricing Prediction (GL4): This project applies data mining techniques to predict house prices in Tunisia. It involves analyzing real estate data, identifying key factors that influence prices, and building predictive models to estimate property values accurately.
# Housing Data Mining Project
A clean, minimal scaffold for end-to-end housing data work (collection, cleaning, feature engineering, modeling, and reporting). It includes a ready-to-run script to merge BNB listing CSVs and a folder layout that keeps raw data separate from derived artifacts.
## Project Structure
```
projet data mining/
├─ data/
│ ├─ raw/ # Original, immutable data dumps (checked in by default)
│ ├─ interim/ # Data in progress of being cleaned/normalized
│ ├─ processed/ # Final, canonical datasets for modeling
│ └─ merged/ # Combined datasets (e.g., merged BNB listings)
├─ notebooks/ # Exploratory analysis, EDA, experiments (Jupyter)
├─ scripts/ # Small utilities and runnable scripts
│ └─ merge_bnb.py
├─ src/ # Reusable Python modules (optional: package later)
│ ├─ data/ # Data loading, cleaning functions
│ ├─ features/ # Feature engineering logic
│ ├─ models/ # Training, evaluation code
│ └─ visualization/ # Plotting utilities
├─ reports/
│ └─ figures/ # Generated graphics and figures
├─ configs/ # Config files (YAML/JSON/env)
├─ docs/ # Project documentation (design notes, decisions)
├─ tests/ # Automated tests
├─ .gitignore # Ignores venv, caches, and generated data outputs
├─ requirements.txt
└─ README.md
```
## Quickstart
1) Create and activate a Python virtual environment (Windows PowerShell):
```powershell
cd "c:\Users\Dell\projet data mining"
.\.venv\Scripts\Activate
```
2) Install dependencies:
```powershell
pip install -r requirements.txt
```
## Data Locations
- Put new raw CSVs in: `data/raw/`
- Generated artifacts are written to: `data/interim`, `data/processed`, `data/merged`, and `reports/figures` (these are ignored by Git by default).
## Data Merging Scripts
The project includes robust merge utilities that handle messy CSVs with embedded newlines and mixed encodings.
### Merge BNB Listings
Combines BNB Ariana, BNB Ben Arous, and BNB Tunis into …