Comparing the Effectiveness of Machine Learning Algorithms in Classifying Google Alerts about Distributed Denial of Service

Distributed denial of service (DDoS) attacks are attempts to make computer or network resources unavailable to its intended users. They cause firms and other organizations significant economic and reputational harm and have risen in frequency and strength over the course of the past years. In order to contribute to the understanding of DDoS attacks, this study explores machine learning as a tool to classify Google alerts about DDoS. I try to answer which machine learning algorithms can improve and simplify the process of retrieving news reporting a DDoS event. Several machine learning algorithms are tested on a dataset and compared in terms of effectiveness. I find the multinomial Naive Bayes algorithm with the bag of words model to be the most effective out of the ones I tested. Furthermore, I explore some applications for the Word2vec algorithm to provide information about semantic features of DDoS related Google alerts.