EML FEXIC Enhanced Machine Learning Model with Automated Feature Extraction for Identification and Classification of Whatsapp Encrypted Traffic

Abstract

In the digital era dominated by encryption and rapid communication, newlineaccurately classifying network traffic has become crucial for network administrators newlineand security professionals. WhatsApp, among the leading encrypted messaging newlineservices, poses unique challenges for traffic analysis. To address these challenges, this newlinestudy introduces the EML-FEXIC framework (Enhanced Machine Learning Model newlinewith Automated Feature Extraction for Identification and Classification of WhatsApp newlineTraffic), employing transfer learning principles for WhatsApp network traffic analysis. newlineThe EML-FEXIC framework signifies a shift in network traffic analysis newlineby utilizing transfer learning to complement traditional machine learning approaches. newlineBy utilizing features extracted from a pre-trained Deep Neural Network (DNN), the newlineframework eliminates the need for extensive manual feature engineering, resulting in newlinemore efficient WhatsApp traffic classification. The DNN is trained using real-time newlinetraffic data collected through port mirroring and benchmark datasets such as ISCX newlineVPN. Additionally, a new dataset was meticulously constructed via port mirroring in newlinePCAP format to capture comprehensive WhatsApp network traffic patterns newline

Description

Keywords

Citation

item.page.endorsement

item.page.review

item.page.supplemented

item.page.referenced