Portfolio

Overview:

This project focuses on predicting lunch box sales using historical data to forecast future demand. The system extracts key features from the data and uses both Linear Regression and Random Forest models to make predictions. The evaluation is based on model performance using RMSE (Root Mean Squared Error) as a validation metric.

Screenshots

Key Features:

1. Linear Regression Model:

The model is designed to predict lunch box demand based on linear relationships between extracted features such as sales trends, day of the week, and seasonal effects.

2. Random Forest Model:

A more complex model that uses an ensemble of decision trees to capture non-linear relationships and interactions between features, providing a more robust prediction for lunch box sales.

3. Feature Extraction:

Key features like historical sales data, weather conditions, holidays, and other external factors are extracted to improve prediction accuracy.

4. RMSE Validation:

The performance of both models is validated using RMSE (Root Mean Squared Error), which measures the differences between predicted and actual sales, ensuring the models’ accuracy.

Category:

Tags:

Links:

• GitHub