This project aims to build a machine learning model that predicts a developer's salary based on features derived from the Stack Overflow Developer Survey dataset.
To develop a machine learning model that predicts an individual's salary using data from the Stack Overflow Developer Survey. The goal is to understand which features (e.g., job title, education level, technologies used) contribute most significantly to salary and to build a regression model for accurate predictions.
- Source: Stack Overflow Developer Survey
- The dataset includes information such as:
- Job Role
- Years of Experience
- Education Level
- Technologies Used
- Country
- Salary (
ConvertedSalary)
- Python
- Pandas
- NumPy
- Scikit-learn
- Matplotlib / Seaborn (for visualization)
- Category Encoders (for encoding categorical data)
- Model Used: Random Forest Regressor
- R² Score: ~60%
- Key Features:
YearsCodePro,Country,EdLevel,LanguageWorkedWith
- Random Forest showed promising results for predicting salaries.
- Top predictive features include professional coding experience, country, and education level.
- R² score indicates a moderate fit, with room for improvement through hyperparameter tuning and feature engineering.
- Develop a GUI/Web app for salary prediction.
- Explore deep learning models using Keras or TensorFlow.
- Improve performance with hyperparameter tuning and feature selection.
- Deploy the model using Streamlit or Flask for user interaction.
- Clone the repository:
git clone https://github.com/Rohit2303A510J0/ADM_PROJECT.git
cd ADM_PROJECT- Install required packages
pip install -r requirements.txt- Run notebook:
jupyter notebook ADM_PROJECT/Salary_prediction.ipynb