Deep Learning Techniques for Assessing Quality of Online Health Articles | Funder: DSERU Interim Funding Investigators: Peter Smith Supervisor: Ashad Kabir, Michael Bewong Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: This project aims to apply various deep learning techniques to automatically assess the quality of online health articles based on 10 quality criteria (proposed in state-of-the-art research). The expected outcome of this project is a tool as a chrome extension that will aid users in shaping their opinion to make the right choice while picking online health-related articles. |
Agriculture Cybersecurity | Funder: DSERU Interim Funding Investigators: Asim Ozan Aras Supervisor: Arash Mahboubi Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: With precision agriculture and smart farms being powered by technology, systems, and data, this project's goal is to investigate cyber risks associated with farms and take proactive steps to mitigate them. |
Deep learning to track the human upper-limb movement for people living with disabilities and stroke | Funder: DSERU Interim Funding Investigators: Sabahat Gulrez Supervisor: Azizur Rahman Ashad Kabir Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: Reconstruction of human upper-body w.r.t the sensors is a challenging task, which will be explored and researched in this project through artificial intelligence. This project will explore the applicability and feasibility of magnetometers or gyros used in mobile phones to track human body and use of computer vision through built in mobile cameras to calibrate the system |
A Lightweight Approach to 3D Measurement of Chronic Wounds | Funder: DSERU Interim Funding Investigators: Kris Simpson Supervisor: Ashad Kabir, Lihong Zheng Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: This research proposal aims to investigate the possibility of developing a lightweight method of performing 3D measurement of chronic wounds, such as ulcers, from high quality 2D video allowing for a distributable or remote assessment of ulcers and other chronic wounds without the need for invasive assessment, or possibly even without the need for the patient to be present in person. |
Use of machine learning in prediction of software vulnerability exploits | Funder: DSERU Interim Funding Investigators: Hadi Eskandari Supervisor: Michael Bewong, Sabih Rehman Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: This research will investigate the inconsistencies associated with the previously developed machine learning models that address discovered software vulnerabilities and identify strategies to improve their performance and reliability. It aims to develop a rigorous evaluation baseline, and further develop strategies that can improve the consistency and effectiveness of the existing machine learning models. |
Analysis of variability of non-nuclear pavement density testing devices | Funder: DSERU Interim Funding Investigators: Noriko Wood Supervisor: Azizur Rahman, Ryan Ip Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: The objective of this project is to determine if Pavement Quality Indicator (PQI) devices are a viable alternative to nuclear density gauges for measuring asphalt density. The use of nuclear density gauges for measuring the degree of compaction of newly constructed asphalt roads contain a radioactive component which may present safety and environmental hazard. Their replacement with non-nuclear PQI devices would eliminate this hazard, which would be a significant safety improvement for workers in the asphalt industry. |
Livestock Supply chain security issues | Funder: DSERU Interim Funding Investigators: Anmol Nayyar Supervisor: Lihong Zheng, Ashad Kabir Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: This research will explore the security issues in the livestock supply chain and provide some recommendations to address those safety issues. During the COVID-19 pandemic, supply chains are being impacted by global lockdowns and data breaches are happening more often than before. Hence, this research will provide insights into how the industry faced the challenges of COVID-19. |
IPv6 hierarchical addressing for large-scale processing, farming and industrial implementation | Funder: DSERU Interim Funding Investigators: David Cullinan Supervisor: Lihong Zheng, Arif Khan Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: As we increasingly see trends in IoT, smart-farming, AI and other technological advancements in traditionally low-tech industries such as agriculture, the purpose of this research proposal is to investigate the efficacy of utilising the incredibly vast address range (in excess of 3.4 x 1038 unique addresses) in order to not only re-think the way that these products may be connected, but also how IPv6 addressing schemes may potentially be assigned. |
A survey on incremental learning | Funder: DSERU Interim Funding Investigators: Bijani Bhasima Supervisor: Michael Bewong, Ryan Ip, Md Geaur Rahman, Zahid Islam Period: 2021 - 2022 Funding Amount: $3,000 Outcomes: Incremental learning often processes the data batch by batch and takes an account of dynamic changes on data distribution and class attributes. Usually, once models are updated, the used data are discarded. Incremental learning thus has a huge significance where all the companies are encountering big data and looking for utilizing them. |