Research Article

Preprocessing Method for Encrypted Traffic Based on Semisupervised Clustering

Table 4

Subset of features in DBSCAN.

Feature descriptionCategory

Proportion of in-payload size in [10, 15) kB to the totalNumerical
Proportion of in-payload size in [15, 20) kB to the totalNumerical
Proportion of in-payload size in [20, 30) kB to the totalNumerical
Proportion of in-payload size in [0, 500) kB to the totalNumerical
Proportion of in-payload size in [500, 1500) kB to the totalNumerical
Proportion of in-payload size in [1500, 2000) kB to the totalNumerical
Proportion of out-payload size in [5, 10) kB to the totalNumerical
Proportion of out-payload size in [20, 30) kB to the totalNumerical
Proportion of out-payload size in [50, 100) kB to the totalNumerical
Proportion of out-payload size in [300, 400) kB to the totalNumerical
Proportion of payload ratio in [5, 10) to the totalNumerical
Proportion of payload ratio in [10, 20)to the totalNumerical
Proportion of payload ratio in [30, 40) to the totalNumerical
Proportion of payload ratio in [150, 200) to the totalNumerical
Proportion of payload ratio in [200, 400) to the totalNumerical
Proportion of payload ratio in [500, 1000) to the totalNumerical
If the communication frequency in [20, 40)Boolean
If the communication frequency in [40, 60)Boolean
If the communication frequency in [70, 80)Boolean
If the communication frequency in [100, 150)Boolean
If the communication frequency in [200, 300)Boolean
If the communication frequency in [400, 500)Boolean
If the connected hosts in [3, 6)Boolean
If the connected hosts in [9, 12)Boolean
If the connected hosts in [15, 20)Boolean