Automatic network problem detection occurring in the campus computer network

In recent years, the use of the internet has become widespread with developing technologies. Internet is used for many needs, especially social media. Today, internet is also needed for remote use of electronic devices used in homes and offices. Continuous access to the internet has become very important for people's quality of life. In this thesis, a method has been developed for early detection of basic faults that may occur in network devices providing internet access and in the lines of these devices. The developed method was carried out in the university campus environment. With this software developed, certain data are taken in real time from network devices. This raw data is pre-processed and made suitable for analysis. Machine learning approaches have been used for this purpose. The data labeled with the expert opinion were used for training, then the data received were classified by Decision Trees, Support Vector Machine and Naive Bayes machine learning methods. Eliminating the malfunctions that may occur in the Internet line requires long time and workload. Thanks to this developed method, maintenance studies are performed by providing early detection of failures that may occur in the university campus environment.