Topic Modeling



No file uploaded.Loading data...
Model loaded successfully.
KMeans model fitted successfully.

First 10 Keywords:
['root', 'user', 'login']

Nearest keyword to cluster centroid:
('login', 0.4863611088756177)
('root', 1.950376979056818)
('root', 0.8207347595313516)
('root', 1.8619936875255354)
('user', 2.047148272817664)

Top TF-IDF scores on each cluster:
Cluster 0:
(authentication,1.35) (euid,1.35) (failure,1.35) (logname,1.35) (nodevssh,1.35) (rhost,1.35) (ruser,1.35) (tty,1.35) (session,1.28) (eventtemplate,1.00)
Cluster 1:
(opened,0.71) (session,0.71) (authentication,0.33) (euid,0.33) (failure,0.33) (logname,0.33) (nodevssh,0.33) (rhost,0.33) (root,0.33) (ruser,0.33)
Cluster 2:
(root,3.00) (check,1.15) (pass,1.15) (unknown,1.15) (closed,0.71) (session,0.71)
Cluster 3:
(abnormally,0.58) (alert,0.58) (check,0.58) (exited,0.58) (pass,0.58) (unknown,0.58)
Cluster 4:
ValueError: max_df corresponds to < documents than min_df