Skip to main content

Table 1 Results of cluster quality evaluation metrics

From: An unsupervised method for social network spammer detection based on user information interests

Dataset

Features

Silhouette coefficient

Davis Bouldin Score

Social Honeypot

GOSS + LOSS

0.133783289

3.619251614

GOSS + LOSS + Entropy

0.185440031

3.878652971

GOSS + Entropy

0.084125872

3.624939240

LOSS + Entropy

0.148410937

3.547088595

The Fake Project

GOSS + LOSS

0.171869530

3.244455815

GOSS + LOSS + Entropy

0.170521355

3.167608666

GOSS + Entropy

0.171822136

3.125462915

LOSS + Entropy

0.180362480

2.941402933

HSpam14

GOSS + LOSS

0.165421365

2.69321541

GOSS + LOSS + Entropy

0.158961042

2.73569256

GOSS + Entropy

0.129411538

2.51342332

LOSS + Entropy

0.172345361

2.49823176