Diet & Exercise Classification using Machine Learning to Predict Obese Patient’s Weight Loss

Obesity-related diseases such as coronary heart disease, stroke, respiratory disorders, etc. has steadily risen in the world over the last decades. Various studies related to obesity have been done; however, there is still a need to predict the possibility of losing obese patient’s weight based on history of his/her diet and exercise data. Therefore, this study use an obese patient as the case study. Diet and exercise data was collected using Smartwatch. This study classifies the obese patient’s level of possibility to lose weight to high (Good health), medium (Normal) and low (Poor health) from the patient's diet and exercise data. Machine learning techniques such as k-nearest neighbour and decision tree are used in this study to classify the diet and exercise data and find out the level of possibility to reduce weight. Analysis of this study shows that the decision tree provides the best accuracy for diet and exercise data where it is recorded 71.54% and 63.63% respectively. On the other hand, k-nearest neighbour shows the accuracy of 65.85% for diet and 69.32% for exercise data. The prediction results of this study can be used by the doctors and physicians to provide better advice and prescription for the obese patients.