Abstract:
"The project, titled ""FitFare Personalized Fitness Meal Plan Application,"" addresses the growing need for individualized meal planning in fitness applications by incorporating cultural preferences, specifically focusing on Sri Lankan cuisine. Many existing fitness apps lack personalization, offering generic advice based on calories, BMI, and activity levels. FitFare aims to bridge this gap by using a machine learning-driven approach that tailors meal recommendations to the user's health data, such as age, height, weight, gender, and activity level, while considering macronutrient balance and cultural relevance.
The solution includes a hybrid algorithm trained using a dataset comprising real and synthetic data enriched with Sri Lankan culinary traditions. This dataset incorporates detailed macronutrient profiles of local foods. The application evaluates users’ fitness goals, preferences, and cultural factors to generate personalized meal plans, ensuring both nutritional accuracy and cultural fit.
Testing the model involved various regression metrics, such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-Squared (R²), which demonstrated high accuracy in meal recommendations. The approach enhances user experience by focusing on long-term health goals, promoting cultural sensitivity, and offering customized nutrition guidance, positioning FitFare as an innovative solution in personalized fitness technology.
This project presents a significant contribution to the fields of health informatics and machine learning, offering a unique blend of cultural awareness and technological innovation to improve personal health outcomes."
