CNN Network Intrusion Detector (KDDCup'99)

Enter the 41 features of a network connection record to determine if it is a Normal connection or an Anomaly (Attack). This model is a 1D Convolutional Neural Network (CNN) optimized for high Attack Recall (using a prediction threshold of 0.4).
Default values are set for a NORMAL FTP data connection.

duration (float, sec)

protocol_type

service

flag

src_bytes (int)

dst_bytes (int)

land (binary)

wrong_fragment (int)

urgent (int)

hot (int)

num_failed_logins (int)

logged_in (binary)

num_compromised (int)

root_shell (binary)

su_attempted (binary)

num_root (int)

num_file_creations (int)

num_shells (int)

num_access_files (int)

num_outbound_cmds (int)

is_host_login (binary)

is_guest_login (binary)

count (float)

srv_count (float)

serror_rate (float)

srv_serror_rate (float)

rerror_rate (float)

srv_rerror_rate (float)

same_srv_rate (float)

diff_srv_rate (float)

srv_diff_host_rate (float)

dst_host_count (float)

dst_host_srv_count (float)

dst_host_same_srv_rate (float)

dst_host_diff_srv_rate (float)

dst_host_same_src_port_rate (float)

dst_host_srv_diff_host_rate (float)

dst_host_serror_rate (float)

dst_host_srv_serror_rate (float)

dst_host_rerror_rate (float)

dst_host_srv_rerror_rate (float)

Attack Probability