Encrypted Network Traffic Analysis and Classification Utilizing Machine Learning.

Encryption is a fundamental security measure to safeguard data during transmission to ensure confidentiality while at the same time posing a great challenge for traditional packet and traffic inspection. In response to the proliferation of diverse network traffic patterns from Internet-of-Things devices, websites, and mobile applications, understanding and classifying encrypted traffic are crucial for network administrators, cybersecurity professionals, and policy enforcement entities. This paper presents a comprehensive survey of recent advancements in machine-learning-driven encrypted traffic analysis and classification. The primary goals of our survey are two-fold: First, we present the overall procedure and provide a detailed explanation of utilizing machine learning in analyzing and classifying encrypted network traffic. Second, we review state-of-the-art techniques and methodologies in traffic analysis. Our aim is to provide insights into current practices and future directions in encrypted traffic analysis and classification, especially machine-learning-based analysis.

Adversarial-Aware Deep Learning System Based on a Secondary Classical Machine Learning Verification Approach.

Deep learning models have been used in creating various effective image classification applications. However, they are vulnerable to adversarial attacks that seek to misguide the models into predicting incorrect classes. Our study of major adversarial attack models shows that they all specifically target and exploit the neural networking structures in their designs. This understanding led us to develop a hypothesis that most classical machine learning models, such as random forest (RF), are immune to adversarial attack models because they do not rely on neural network design at all. Our experimental study of classical machine learning models against popular adversarial attacks supports this hypothesis. Based on this hypothesis, we propose a new adversarial-aware deep learning system by using a classical machine learning model as the secondary verification system to complement the primary deep learning model in image classification. Although the secondary classical machine learning model has less accurate output, it is only used for verification purposes, which does not impact the output accuracy of the primary deep learning model, and, at the same time, can effectively detect an adversarial attack when a clear mismatch occurs. Our experiments based on the CIFAR-100 dataset show that our proposed approach outperforms current state-of-the-art adversarial defense systems.

An automatic medication self-management and monitoring system for independently living patients.

This paper describes the development, prototyping, and evaluation of RMAIS (RFID-based Medication Adherence Intelligence System). Previous work in this field has resulted in devices that are either costly or too complicated for general (especially elderly) patients to operate. RMAIS provides a practical and economical means for ordinary patients to easily manage their own medications, taking the right dosage of medicine at the prescribed time in a fully automatic way. The system design has the following features: (1) fully automatic operation for easy medication by using the built-in scale for dosage measurement and a motorized rotation plate to deliver the right medicine container in front of a patient, (2) various medication reminder messages for patients, and noncompliance alerts for caregivers (such as doctors, relatives or social workers who take care of the patients), and (3) incremental and economical adoption by pharmacies, patients, and insurance companies.

RMAIS: RFID-based medication Adherence Intelligence System.

There has been compelling evidence that outpatients, especially those who are elderly or taking multiple complexly scheduled drugs, are not taking their medicines as directed, leading to unnecessary disease progression, complications, functional disabilities, lower quality of life, and even mortality. Existing technologies for monitoring and improving drug adherence are either costly or too complicated for general patients to use. In this paper, we introduce the detailed design and the complete prototype of a marketable Radio-Frequency Identification (RFID)-based Medication Adherence Intelligence System (RMAIS) that can be conveniently used at a residential home by ordinary patients. RMAIS is designed to maintain patients' independence and enable them to take multiple daily medicine dosages of the right amount at the right time. The system is patient-centered and user-friendly by reminding a patient of the prescribed time for medication and dispensing it in a fully automatic and fool-proof way. This is achieved mainly due to its novel design of a motorized rotation platform and the smooth integration of a scale, an RFID reader, and the rotation platform. In addition, this system has an Internet-based notification function that is used to alert the patient when it is time to take medicine as well as report deviations from the prescribed schedule to the primary care physicians or pharmacists.