Development and Evaluation of an Attack Detection System in a Computer Network
This project saw the implementation and evaluation of an Anomaly Network Intrusion Detection System (ANIDS) based on soft computing techiniques such as genetic algorithms and fuzzy logic. The system was tested with UGR’16 dataset and compared with several other consolidated techniques such as kNN, SVM, logistic regression, naive Bayes and LSTM neural networks. Various scenarios have been designed to check the behaviour of the implemented systems. The entire project has been realized in Python language using the following libraries: scikit-learn, pandas, Keras, TensorFlow.
This project has been developed as master thesis project in the computer engineering master’s degree program at University of Pisa.