<![CDATA[Acute Respiratory Infection (ARI) is a common and serious health problem worldwide. Early diagnosis of ARTI is crucial for appropriate treatment and disease control. However, ARTI diagnosis requires time, cost, and high expertise. Therefore, an automated method is needed to assist in the ARTI diagnosis process. This research applies the concept of Non-deterministic finite automata (NFA) in the diagnosis of ARTI. NFA is a mathematical model that can describe non-linear and unstructured systems. The aim of this study is to develop an automated method that can identify complex patterns in ARTI symptoms and provide accurate diagnoses. The research method employed in this study is experimental, utilizing NFA implementation using the J-Flap application. The data used consists of ARTI symptom data from previous studies. The NFA design is created based on ARTI symptoms and the possible transitions between these symptoms. Testing is conducted using different ARTI cases. The research results demonstrate that the NFA successfully recognizes patterns in ARTI symptoms and provides diagnoses that align with the input data. The use of NFA in ARTI diagnosis using J-Flap can achieve high accuracy. Furthermore, this research offers improved diagnostic options compared to previous studies. Consequently, this study addresses the challenges posed by the complexity of ARTI symptom variations by applying the NFA concept. The developed automated method can provide accurate diagnoses for ARTI and overcome the limitations of the traditional diagnosis process.]]>
