Over the last few years there has been a notable increase in the extent and impact of network attacks. These attacks aim to compromise the confidentiality, integrity, or availability of data and network resources. Furthermore, decision-making becomes crucial in formulating proactive strategies on prevention or detection tasks in order to respond promptly to these network attacks. Besides, there are many approaches to identifying these attacks and making decisions but machine learning techniques are the most popular and reliable for identifying unknown attackers and achieving complete process automation. In this paper a set of Machine Learning methods is used, in particular boosting algorithms to enhance the attack detection process and to create multiple models and then combine them to produce improved results.