Samer Samarah

Remote recognition of human activities has revolutionized the healthcare industry by enabling smart monitoring of patients at their residents. However, such systems often expose sensitive data, making them vulnerable to breaches. Preserving data confidentiality is a problem that remains underexplored in smart activity recognition, despite its important in preserving patients’ privacy. This paper aims to develop an optimized feature augmentation technique using fractional derivatives to increase data uncertainty for improved confidentiality, while maintaining acceptable classification accuracy. A key challenge in this domain is the balancing of privacy and performance. To achieve this goal, the proposed framework utilizes multilayer perceptron neural networks that are used to embed multiple modalities of data and integrate them into a coherent structure. Validation was performed using five state-of-the-art classification techniques to measure the performance of the proposed framework in terms of classification accuracy and data confidentiality. The results elevate the potential of the proposed methodology to enable quality healthcare services in terms of confidentiality and performance.

One of the serious security breaches that threatens today's smart technologies is the broadcasting of false alarms. These alarms may severely affect road management systems. Vehicular networks suffer from this type of threat because of the opacity and the on-purpose broadcasting of false contents that occur frequently. Modern cloud-fog communication architectures force the organization of hierarchical communications, resulting in more robust governance of disseminated contents. This paper proposes a prediction technique that is based on a binary classification using hidden Markov model (HMM). Our contributions in this research area are as follows: 1) modeling fog data features to enforce organization of communication, 2) proposing a technique for the automatic modeling of HMM parameters, which facilitate the precise prediction of false contents, and 3) proposing a dissemination algorithm for fast communication. We conducted several experiments to measure and compare the performance of the proposed model with that of state-of-the-art classification algorithms. We also performed a misuse analysis to measure the ability of our proposed model to detect false alarms in synthesis and real-world datasets. In addition, we conducted a further analysis to demonstrate the sensitivity of our methodology to cutoff points. The results showed that our HMM-based technique provides promising prediction accuracy and significantly outperforms existing classification techniques in detecting false-alarm contents

Due to the recent advances in vehicular ad hoc networks (VANETs), smart applications have been incorporating the data generated from these networks to provide quality of life services. In this paper, we have proposed taxonomy of data mining techniques that have been applied in this domain in addition to a classification of these techniques. Our contribution is to highlight the research methodologies in the literature and allow for comparing among them using different characteristics. The proposed taxonomy covers elementary data mining techniques such as: preprocessing, outlier detection, clustering, and classification of data. In addition, it covers centralized, distributed, offline, and online techniques from the literature.

Advances in multimedia technologies have led to the emergence of smart home applications. In fact, mobile multimedia technologies provide the infrastructure to adopt smart solutions and track inhabitants' activities. In-home activity recognition significantly enhances the performance of healthcaremonitoring and emergency-control applications for elderly and people with special needs. Developing and validating data models for such applications requires training sets that reflect a ground truth in the form of labeled or annotated data. With the accelerated development of Internet-of-Things applications, automated annotation processes have emerged understanding resident behavior in terms of activities. This paper presents a methodology for automatic data annotation by profiling sensing nodes. Our proposed methodology models activities based on spatially recognized actions, with every activity expected to have a direct relationship with a specific set of locations. Furthermore, the proposed technique validates the assignment of labels based on the temporal relations among consecutive actions. We performed experiments to evaluate our proposed methodology on CASAS data sets, which indicated that the proposed methodology achieved better performance, to a statistically significant extent, than the state-of-the-art methodologies presented in the literature.

Recent advances in wireless sensor networks for ubiquitous health and activity monitoring systems have triggered the possibility of addressing human needs in smart environments through recognizing human real-time activities. While the nature of streams in such networks requires efficient recognition techniques, it is also subject to suspicious inference-based privacy attacks. In this paper, we propose a framework that efficiently recognizes human activities in smart homes based on spatiotemporal mining technique. In addition, we propose a technique to enhance the privacy of the collected human sensed activities using a modified version of micro-aggregation approach. An extensive validation of our framework has been performed on benchmark data sets yielding quite promising results in terms of accuracy and privacy-utility tradeoff.