This project analyzes data from the Divar platform (an advertising company in Iran), including Exploratory Data Analysis (EDA), statistical analysis, recommender system, and price/rent prediction. The main goal is to use machine learning techniques to better understand the data and provide predictive models. 🚀
The dataset used in this project is not available in the repository due to its large size (approximately 1 million records 📈). It consists of 64 columns, including:
- 20 numerical columns 🔢
- 44 categorical columns 🏷️
The divar_project repository is divided into 5 main sections in src/:
- EDA (Exploratory Data Analysis) 🔍: Exploratory analysis of data to understand distributions, relationships, and patterns.
- Statistical_analysis 📊: Advanced statistical analyses such as statistical tests and statistical modeling.
- recommender_system 🤖: Implementation of a recommender system for suggesting products or ads.
- prediction_price 💰 and prediction_rent 🏠: Predictive models for purchase price and rent. These two sections are in a shared folder.
- Data Processing: pandas 📊, numpy 🔢, scipy 🔍
- Visualization: matplotlib 📈, seaborn 🎨, plotly 📊, geopandas 🗺️
- Machine Learning: sklearn 🤖, scipy 🔬
- Algorithms and Models: k-means 📍, DBSCAN 🔄, LightGBM 🌟, Random Forest Regressor 🌲
- Python 3.11 🐍
- Install required libraries via
pip install -r requirements.txt(the requirements.txt file should be available in the repository).
- Clone the repository:
git clone https://github.com/username/divar_project.git📥 - Navigate to the project directory:
cd divar_project📁 - Create a virtual environment (optional):
python -m venv env🏗️ - Install libraries:
pip install -r requirements.txt📦 - For each section, run the corresponding scripts (e.g., for EDA:
python eda/main.py▶️ ).
- The original data is not uploaded to the repository due to its size. Please download the data from the relevant source and place it in the
data/folder. 💾 - For questions or collaboration, use Issues or Pull Requests. 💬
This project is released under the MIT License. 📄