TY  - BOOK
AU  - 	Daroya,Renz C.; Dela Cruz,Donnabel C.
TI  - Adaptation of fitness application through AI integration: An android based application for home workout fitness experience
AV  - T60 D37 2025 
U1  - . 
PY  - 2025///
CY  - .
PB  - .
KW  - academic writing
N1  - ABSTRACT: The “iFit AI” fitness application is a user-friendly platform created to assist gym beginners, users who prefer to workout at home, gym enthusiasts, and gym instructors in determining the suitable workout for different body types. Many beginners struggle to because they lack guidance on workouts, diet, and exercise intensity. This study aims to bridge that gap by utilizing predictive analytics to offer personalized workout routines and diet recommendations. By analyzing a user’s past performance, physical attributes, and other relevant data, the app can forecast the outcomes of specific workouts, leading to tailored exercise and nutrition plans. These customized approaches enhance fitness efficiency, minimize injury risk, and promote overall health. Additionally, a progress tracker will motivate users by showing their achievements toward their fitness goals. The developers will use Multi-Linear Regression (MLR) to predict results based on users consistent efforts, providing actionable insights for workout and diet adjustments. Users will also have the option to upload photos for visual progress tracking using TensorFlow. To boost engagement, the app will feature a rewards system that incentivizes users upon reaching goals while recommending appropriate food options to align with their fitness objectives. The approach has been successful in correctly identifying different proper workouts and calorie intake for each body type. iFit AI has proved its ability to appropriately recommended and predict through thorough testing. This achievement demonstrates TensorFlow’s usefulness in this setting while also validating the system’s design and implementation. While effective, the researchers believe that utilizing algorithms created specifically for complicated visual data could result in even greater accuracy; F
ER  -