Development of expert system for skin diseases based on named-entity recognition (NER) and fuzzy inference

The skin is an important part of the human body to be cared for. In healthy conditions, the skin has an important role in protecting the body from various diseases and infections. On the other hand, the skin is also susceptible to diseases caused by viruses such as measles, German measles, and chickenpox. In that case, an expert system to detect measles, German measles, and chicken pox is required. Fuzzy Inference has the ability to give flexible recommendations in the expert system. Named-entity recognition (NER) can detect symptom entities in text input that is inputted by the user in the form of natural language. In this study, we apply fuzzy inference and NER to develop an accurate expert system. The NER model, which obtained the best F1 score of 0.83, was able to detect entity symptoms with a very good level of accuracy using the ALBERT model, according to the data presented above. A web-based system that is very user-friendly has been developed as a result of the success of this trial, making it simpler for users to enter symptoms of children's skin conditions. Based on symptom data that is entered in the form of free text, this system's implementation has demonstrated its capacity to offer a diagnosis of children's skin disease