Acute Respiratory Infections Diagnosis Using Learning Vector Quantization

Acute Respiratory infections are a disease that is quite widely suffered throughout the world, moreover, every year in the world 13 million children under 5 years old die 95% of them are from developing countries and one-third die due to acute respiratory infection. based on a report from the Ministry of Health in 2020 There are 34.8% of children suffering from this disease in Indonesia. Many people do not understand the symptoms that arise when suffering from ARI, especially if the incident happened at home without medical assistance. For this problem, a system is needed that can help people to diagnose acute respiratory infections. This study implements the learning vector quantization method to diagnose the symptoms experienced whether they belong to the classification of common cold and flu, asthma, pneumonia, or tuberculosis, data that has been entered by the user will be compared with previously created models using a training dataset. As for the testing in this study using the classification accuracy method and from the results of this study, the accuracy of the best method was obtained at 97.50% at split validation of 80:20 and an average accuracy of 85.92%.