| dc.description.abstract | Anemia is still one of the most common health problems among adolescents, especially among adolescent girls. To avoid more serious impacts, a fast and precise detection system is needed. This research aims to develop a machine learning-based anemia status classification system, as well as provide food suggestions that are in accordance with individual conditions. The classification model used is Support Vector Machine (SVM), with input data coming from the results of filling out questionnaires by teenagers, including automatic calculation of Body Mass Index (BMI) based on weight and height. The data was divided with a proportion of 80% for training and 20% for testing. The model evaluation results showed an accuracy of 95.12%, with a precision value of 0.94, recall of 0.94, and f1-score of 0.95. Testing using the 5-fold cross-validation method resulted in an average accuracy of 96.97%. As an additional feature, the system is equipped with a food recommendation module that uses the Content Based Filtering (CBF) approach. This method applies the Dot Product technique between the user's nutritional needs and the nutritional composition of food ingredients, such as iron, protein, energy, and vitamin C. The nutrient content data is taken from the Food Composition Table. Nutrient content data is taken from the Indonesian Food Composition Table (TKPI). Recommendations are given based on the user's anemia classification (Severe Anemia, Mild Anemia, and No Anemia), with a list of food ingredients in both raw and processed forms. The findings of this study show that the integration between classification methods and recommendation systems has potential as a digital solution in monitoring and managing anemia in adolescents. | en_US |
