Neural approach can be used for the detection of cyber attacks. Confidentiality of data due to the unauthorized access can cause compromises over the user's system. Cyber Attacks are prominently increasing in recent years where the users lose their confidentiality and privacy over the data. IoT is used nowadays everywhere in this digital era. For the classified approach and for detecting the network intrusion deep neural approach is implemented. IoT applications are used in industrial and technological sectors. Determine the dataset with accurate parameters to get precise result over the detection of cyber attacks. Generally, the network traffic can be reduced by using neural approach rather than machine learning approach. The increase in the cyber attacks in IOT field has to be identified to avoid the breach over cyber physical system (cps). Some of the cyber attacks can be detected by anomaly detection. The intruder tries to attack the victim by explicating the protocols and websites. Define the attack by classifying the network approach using dnn. The back end of the applications gets compromised resulting data confidentiality and privacy breach over data. The malicious activity can be ensemble by an Ann approach. To determine the cyber attacks in an advanced way, hypo testing approach has been implemented