• Description:

This is the data set used for The Third International Knowledge Discovery and Data Mining Tools Competition, which was held in conjunction with KDD-99 The Fifth International Conference on Knowledge Discovery and Data Mining. The competition task was to build a network intrusion detector, a predictive model capable of distinguishing between 'bad' connections, called intrusions or attacks, and 'good' normal connections. This database contains a standard set of data to be audited, which includes a wide variety of intrusions simulated in a military network environment.

Split Examples
'test' 311,029
'train' 4,898,431
  • Feature structure:
    'count': int32,
    'diff_srv_rate': float32,
    'dst_bytes': int32,
    'dst_host_count': int32,
    'dst_host_diff_srv_rate': float32,
    'dst_host_rerror_rate': float32,
    'dst_host_same_src_port_rate': float32,
    'dst_host_same_srv_rate': float32,
    'dst_host_serror_rate': float32,
    'dst_host_srv_count': int32,
    'dst_host_srv_diff_host_rate': float32,
    'dst_host_srv_rerror_rate': float32,
    'dst_host_srv_serror_rate': float32,
    'duration': int32,
    'flag': ClassLabel(shape=(), dtype=int64, num_classes=11),
    'hot': int32,
    'is_guest_login': bool,
    'is_hot_login': bool,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=40),
    'land': bool,
    'logged_in': bool,
    'num_access_files': int32,
    'num_compromised': int32,
    'num_failed_logins': int32,
    'num_file_creations': int32,
    'num_outbound_cmds': int32,
    'num_root': int32,
    'num_shells': int32,
    'protocol_type': ClassLabel(shape=(), dtype=int64, num_classes=3),
    'rerror_rate': float32,
    'root_shell': bool,
    'same_srv_rate': float32,
    'serror_rate': float32,
    'service': ClassLabel(shape=(), dtype=int64, num_classes=71),
    'src_bytes': int32,
    'srv_count': int32,
    'srv_diff_host_rate': float32,
    'srv_rerror_rate': float32,
    'srv_serror_rate': float32,
    'su_attempted': int32,
    'urgent': int32,
    'wrong_fragment': int32,
  • Feature documentation:
Feature Class Shape Dtype Description
count Tensor int32
diff_srv_rate Tensor float32
dst_bytes Tensor int32
dst_host_count Tensor int32
dst_host_diff_srv_rate Tensor float32
dst_host_rerror_rate Tensor float32
dst_host_same_src_port_rate Tensor float32
dst_host_same_srv_rate Tensor float32
dst_host_serror_rate Tensor float32
dst_host_srv_count Tensor int32
dst_host_srv_diff_host_rate Tensor float32
dst_host_srv_rerror_rate Tensor float32
dst_host_srv_serror_rate Tensor float32
duration Tensor int32
flag ClassLabel int64
hot Tensor int32
is_guest_login Tensor bool
is_hot_login Tensor bool
label ClassLabel int64
land Tensor bool
logged_in Tensor bool
num_access_files Tensor int32
num_compromised Tensor int32
num_failed_logins Tensor int32
num_file_creations Tensor int32
num_outbound_cmds Tensor int32
num_root Tensor int32
num_shells Tensor int32
protocol_type ClassLabel int64
rerror_rate Tensor float32
root_shell Tensor bool
same_srv_rate Tensor float32
serror_rate Tensor float32
service ClassLabel int64
src_bytes Tensor int32
srv_count Tensor int32
srv_diff_host_rate Tensor float32
srv_rerror_rate Tensor float32
srv_serror_rate Tensor float32
su_attempted Tensor int32
urgent Tensor int32
wrong_fragment Tensor int32
  • Citation:
@misc{Dua:2019 ,
  author = "Dua, Dheeru and Graff, Casey",
  year = 2017,
  title = "{UCI} Machine Learning Repository",
  url = "http://archive.ics.uci.edu/ml",
  institution = "University of California, Irvine, School of Information and
Computer Sciences"