Cyber-attack detection by an intrusion detection system (IDS) in web mining involves utilizing the IDS to identify and mitigate malicious activities and threats specifically targeting web mining operations. The IDS monitors the network traffic associated with web mining operations. It analyses the data packets exchanged between the web mining server and clients, looking for any suspicious or anomalous patterns. This IDS combines the principles of intrusion detection with the analysis of web data to identify potential threats or anomalies. In this article, the IDS is implemented through Continuous Learning Vector Quantization (CLVQ) algorithm for identifying and classifying the intrusions presented in web based systems. Also, the Association Rule Mining (ARM) for similarity determination is introduced. Based on the experimental results, such pre-processing and combination of similarity determination method are applied with CLVQ classification algorithm in machine learning models using KDD’ 99 dataset. The proposed CLVQ classification achieves higher classification accuracy.