Enhancing FTP Security Through Ensemble Learning-based Brute Force Attack Detection

A common network environment, File Transfer Protocol (FTP), efficiently transfers data between clients and servers on a computer network. FTP, a client-server protocol, simplifies file administration by uploading and downloading files. FTP allows clients to request and transmit files to servers, facilitating smooth data interchange. FTP is simple but sends data in plain text, hence secure variations like FTPS and SFTP were developed to solve encryption and security. Network risks must be detected early to avoid harm and illegal access. Finding security risks early on lets you take quick and preventative steps, which lowers the chances of data breaches, system compromises, and network interruptions. Early detection helps prevent malicious activity from escalating, protecting network integrity and sensitive data. This proactive strategy is essential for network security and resilience against emerging cyber threats. The research work showed that machine learning methods work, especially when it comes to finding brute force attacks on FTP sites. With the help of the CIC_IDS_2018 dataset, the model showed that it was very accurate, showing how well it can spot early risks. It was much easier for the protective system to hold up after its settings were carefully changed. The mathematical models showed the complex behavior, especially the changes in the maximum depth. These results show how useful machine learning can be for making networked systems stronger against certain cyber risks. They also give us important information for creating proactive cybersecurity strategies. Machine learning recognized FTP-bruteforce attacks effectively in the experiment. The algorithm detected and prevented early threats well using the CIC_IDS_2018 dataset.