Preprocessing Method for Encrypted Traffic Based on Semisupervised Clustering

Table 1

Designed feature set.

Feature name	Description	Number
Inbound payload size	Interval granularities with 2, 3, 5, 10, 20, 50, 100, 200, 500, 5000 kB	100
Outbound payload size	Interval granularities with 2, 3, 5, 10, 20, 50,100, 200, 500, 5000 kB	100
In-and-out payload ratio	Interval granularities with 2, 3, 5, 10, 20, 50, 100, 200, 500, 5000	100
Communication frequency	Interval granularities with 2, 3, 5, 10, 20, 50, 100, 200, 500, 5000	100
Connected devices	Interval granularities with 2, 3, 5, 10, ,20, 50, 100, 200, 500, 5000	100