A machine learning project to predict bike rental demand using multiple linear regression analysis.
This project develops a multiple linear regression model to predict demand for bike sharing services.
Domain: Shared Mobility & Transportation
This project focuses on bike sharing systems - a form of sustainable urban transportation where bicycles are made available for shared use on a short-term basis. These systems have become increasingly popular in cities worldwide as an eco-friendly alternative to traditional transportation.
The bike sharing company BoomBikes needs to predict daily bike rental demand to:
- Optimize operations and resource allocation
- Maximize profitability
- Better understand factors affecting demand patterns
- Plan for post-pandemic recovery strategies
The analysis uses bike-sharing data from US bike-sharing provider BoomBikes, containing information about:
- Daily bike rentals for casual and registered users
- Weather conditions
- Seasonal patterns
- Holiday information
- Temperature and other environmental factors
-
Key Predictor: The variable "registered" has the highest correlation (0.945) with the target variable "count"
-
Numerical Impact: Variables "temp" and "year" have the highest coefficients, indicating the strongest impact on rental patterns
-
Categorical Factors: Season, weather conditions, and month of the year significantly impact rental patterns
-
Model Performance: R² score of test set is 0.82 compared to train set 0.78, indicating the model generalizes well without overfitting
- Python: 3.13.5
- NumPy: 2.3.2
- Pandas: 2.3.1
- Matplotlib: 3.10.5
- Seaborn: 0.13.2
- Scikit-learn: 1.7.2
- Statsmodels: 0.14.5
- SciPy: 1.16.1
Created by @vmeghmalaiiib - feel free to contact me!