The use of information technology and computers has become a necessity in daily work, both in private and public agencies, especially in terms of service to the community, the use of information technology is certainly very supportive of existing activities. One of the activities at the Department of Population and Civil Registration is the large number of incoming and outgoing letters, this causes problems in archiving, both incoming and outgoing letters, so to overcome this a website-based archive system for incoming and outgoing mail is built. This system was built using the waterfall method. The test results on the archive of incoming and outgoing letters based on this website indicate that this application has been running well, this can be seen in the results of the black box testing which shows that the application is running as expected, and makes it easier for employees to carry out their duties of archiving incoming and outgoing mail.

An expert system is a type of artificial intelligence application that is used to tackle complex problems that require specialized knowledge. Expert systems can be used in a variety of disciplines, including healthcare, finance, and manufacturing. The aim of this study was to apply the Nave Bayes approach in a website-based ear illness diagnostic system and to determine its accuracy in an expert system for diagnosing ear disease. The naive Bayes approach is implemented in this research because it may assume that each symptom is independent of one another and can thus be used to assess the probability of a condition based on the symptoms that emerge. The results of this study show that the expert system for diagnosing ear disease using the Nave Bayes method is built on a website using the PHP programming language and the database maintained by MySQL, and this application has been tested 10 times, with 9 test data appropriate and 1 test data not appropriate. As a result of testing this application, the accuracy value obtained is 90%.