My research interests include Intelligent Systems, Computer Networking, Distributed Systems, Machine Learning, Big Data, Serious Games, Multimedia Systems and Communication, Internet of Things, CS Education, and eHealth.
Worldwide universities and colleges have a problem with retaining and graduating students – particularly students in STEM areas. People also argue that the core computer science course fails because it is boring, antisocial and irrelevant. A number of attractive teaching methods have been proposed and tested such as active learning, hybrid learning, social learning and flipped classroom. Most of these do not consider students with varying cognitive skills. The computer science and other departments are trying to discover promising teaching methodologies appropriate for cognitively diverse students so that the department can improve student learning outcomes and retention rate. This project will study a new instrument, serious games – games beyond entertainment, augmented into the current teaching methodologies for the core computer science courses such as data structures and algorithms. People learn in different way, so one-size-fits-all approach to learning would be unproductive. Moreover, this is hard to determine appropriate learning styles for a student for each course. Learners can be grouped into seven learning styles (i.e. visual, aural, verbal, physical, logical, solitary and social). But in practice, learners have a bit of all styles, however some are more predominant. This research project will investigate serious gaming method over a specific course that will incorporate various learning styles together through a game and record its success and/or failure with respect to various objectives such as retention, progression, graduation rate, engagement, faculty-student interaction, feedback, learning outcomes, confidence and applied knowledge. These research will allow us to enhance educational practices so that students benefit directly. The college and university benefits include retention, improved learning outcomes, academic performance and quality enhancement.
In restricted and defined circumstances, surveillance system can bring public safety and detect criminal activities. Many public places including shopping malls, schools, office buildings, and highways are now equipped with multimodal sensors such as cameras, motion sensors, microphones, RFID etc. within the existing surveillance infrastructure. Data from these sensory devices are captured, processed and analyzed in order to detect unusual events, which are then reported to control stations or human operators for determining further preventive actions. Although these existing surveillance infrastructures have proved to be useful from a security perspective, there are two important issues that bring major apprehension among people, which are privacy safeguarding and quality of surveillance information (QoI). Due to the potential of being exposed, many people are increasingly being reluctant to be electronically monitored. On the other hand, due to the low quality of surveillance information, because of noisy and imprecise sensor data, we witness many undesired consequences such as false alarms, service interruptions, and often violation of privacy. It is worth noting that although there has been a significant progress in the field of surveillance research, the issues related to privacy preservation and determination of QoI in public security applications have often been overlooked. This is partially due to the fact that it is relatively difficult to achieve the goals of safety and security, while preserving privacy. To achieve these goals, it is desirable to have a mechanism that can automatically determine in real time what to monitor/record and what to hide/conceal based on the processed QoI, though challenging. This research is aimed to investigate scientific solutions of the above research problems in order to perform effective and automated surveillance.
The growing number of elderly population at home and abroad require improved caring facilities. Elders, often with cognitive and physical impairment, need assistance in their activities of daily living (ADLs), which is usually provided by human caregivers (HCGs). As the demand for caregiver's assistance increases, the shortage of traditional care resources becomes obvious. I am working on to devise a Virtual Caregiver (ViCare) framework that supports a HCG to monitor elderly by being aware of their surroundings. The ViCare system attempts to understand the elderly persons' activities and contexts based on the data captured by the sensors placed in their environment and dynamically decides what services to provide them or whether there is a need to interrupt HCG depending on the type of activities. It will not only minimizes cognitive load of a HCG but also will provide a seamless assistance to the elderly toward their improved health and well-being in their living environment.
I am planning to devise a P2P framework that can survive in a faulty network environment where peers can dynamically join and disappear. The model will construct multi-path communication framework while considering network latency and processing delay. This is challenging especially in a gaming context, because games typically impose time-constraints on the communication layer that must be satisfied to ensure the quality of the applications. I intend to investigate how the cheaters can exploit the P2P paradigm, and to devise a P2P architecture that prevents cheaters from successfully modifying the gameplay for their advantage. A networked virtual environment is a very complex application, and simulation can only go so far in validating algorithms and architectures targeted at such systems.
Obesity is a major health concern that can lead to serious health problems including increased risk of cardiovascular disease, type2 diabetes, and insulin resistance in addition to psychosocial problems, functional limitations, and disabilities that adversely affect the social and physical abilities of obese person in daily affairs. One innovative approach to address this problem is serious games which combine exercise and gaming in more attractive manners. I am working to augment current obesity treatment methods and empower therapists and patients with a novel computing system for continuous treatment and long-term health management of obesity, especially for children and youth. The aim is to develop fundamental technologies and systems using pervasive computing, ambient sensors, and mobile gaming technologies. The system will provide stakeholders with improved tools and techniques to prevent or treat obesity by enabling continuous and accurate measurements of a patient's. At the same time, the system will use serious games to encourage and support the integration of a healthy lifestyle for the youth. The group nature of game will enhance socialization and increase patient’s confidence which is important for obese children.