@thesis{thesis, author={Teguh Arif Hiayatulloh}, title ={Implementasi fuzzy inference system model sugeno deteksi serangan botnet}, year={2020}, url={https://repository.ittelkom-pwt.ac.id/5712/}, abstract={Botnet attacks using DDoS method often times targeting business servers, which caused effect to the servers response from client’s requests slowed down significantly, even stopped server to stop responding client’s requests. The attacks usually targeting a server and caused the traffic to increase drastically so that server unable to respond to any requests. On the other hand, Fuzzy Inference System (FIS) could be used to predict occurrences such as rainfall density, the number of new students, etc with inputting compilation of variables. This FIS method is expected to predict if traffic is from Botnet or not. The characteristic of a Botnet traffic are bytes that exceeding limitation and a streak of Source IP Address that suddenly come in. Sugeno’s Fuzzy Inference System is a derivative from fuzzy model, which its characteristic is in the output. Sugeno FIS’s output not using a compilation of fuzzies, it using a constant or linear equation instead so that it match this project’s output. Sugeno’s FIS process in this research use three input variables and an output variable. Input variables that used is Protocol, Homogenity Source Address and Byte. Meanwhile, the output variable is a label, which divided into two compilations, Backgrounds or Legitimate and Botnet. The orders in this Sugeno’s FIS are Fuzzyfication, reasoning using rule, Deffuzyfication and determine if the output is a Botnet or not using Deffuzyfication. The accuracy of this method when calculated using Confusion Matrix is 90%. According the result, Sugeno’s FIS is suitable for detecting Botnet. The high accuracy of the result means that Sugeno FIS is able to generate an expected respons, which is to distinguish from each variables to detect DDoS Botnet’s attacks. Keywords : Botnet, Confusion matrix, DDoS, Fuzzy Inference System, Sugeno} }