Open AccessDetermining Attributes of Encrypted Data Traffic using Feature Selection Method
Author(s) -
Tasmi Tasmi,
Herri Setiawan,
Deris Stiawan,
Husnawati Husnawati,
Sasut Analar Valiata
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.a2674.109119
Subject(s) - encryption , computer science , feature selection , data mining , ranking (information retrieval) , feature (linguistics) , malware , selection (genetic algorithm) , pattern recognition (psychology) , machine learning , artificial intelligence , computer security , philosophy , linguistics
Encrypted packages such as banking and e-commerce are widely used in various fields because it is advantages in terms of data security. However, the problem occurs when checking attributes package to determine if it is a safe packet instead of malware. The purpose of this study is to get the best attributes using feature selection processes by ranking.The results of this study found that from the two best methods of IG and One R, in average IG better than One R. If based on the results of the response, the data produced for the estimated data of TLS V1.0 IG method has better accuracy compared to the One R method, on the contrary inTLS V1.2 One R data is better than IG.