A Multi-Classification Approach for the Detection and Identification of eHealth Applications

Abstract

eHealth services category has a diversified set of traffic patterns and demands in terms of QoS assurances. Existing QoS solutions were designed to support only aggregated classes of service and cannot differentiate traffic based on an application's behavioral pattern. In order to improve the performance of eHealth applications for home and mobile users there is a need to develop new traffic identification techniques, which would work at the edge of the network. This paper addresses the above problem by proposing machine learning-based approach for eHealth traffic identification. We investigate different techniques which combine the results from multiple machine learning classifiers and show which combination of techniques is best suited for identifying diverse eHealth traffic. Our approach is validated in a mobile e-health application context and the results prove that multi-classification techniques can be used in practice to provide application-based service differentiation.

Measuring the accuracy of open-source payload-based traffic classifiers using popular Internet applications

“…While open-source traffic classifiers can be easily downloaded and used for this purpose, there is little published evidence that the classifications produced are accurate and reliable. L7 Filter [3], in particular, frequently appears in recent traffic classification literature, either as a source of signatures that were used to develop the classification engine [4] or to provide ground truth during validation [5] [6] [7], despite earlier studies suggesting that L7 Filter has a high rate of misclassification [8] [9].…”

SEAIoT: Scalable E-Health Architecture based on Internet of Things

“…These mobile e-health applications can be improved by providing a multi-classification technique to determine traffic generated by its users. The researchers on [6] found that this solution can be applied to other applications that are not related to e-health. Furthermore, time, complexity, and efficiency are not measured.…”

“…Monika Grajzer et al, [2] presented one of the recent machine learning-based techniques for identifying and detecting traffic on mobile e-health applications. These mobile e-health applications can be improved by providing a multi-classification technique to determine traffic generated by its users.…”

Online Traffic Measurement and Analysis in Big Data: Comparative Research Review

“…Historically, the L7 Filter signatures have been popular within the traffic classification community. During the same period, researchers requiring a free deep packet inspection Dots Per Inch (DPI) tool to provide ground truth data for testing and evaluating classification techniques, found that L7 Filter was the only feasible option (Grajzer et al, 2012;Dong et al, 2013;Carela-Espanol et al, 2011).…”