Advanced Topics in Information Systems Report a Broken Link

COMP 694: Advanced Topics in Information Systems is a three-credit, independent study course that is designed to provide SCIS graduate students with a variety of learning opportunities. All graduate students have access to COMP 694 and are encouraged to attend its seminars, but only MScIS students can register for credit.

This course serves to update, enrich, and diversify your knowledge of information systems. It encourages you to engage with faculty members and prepare yourself in choosing a route, topic, and supervisor for your final research. COMP 694 provides opportunities for you to explore areas of interest not provided in the core curriculum or elective courses.

In this course, you will attend various academic seminars on the latest research advancements and technology developments. You will summarize and reflect upon the seminar topics that interest you the most. You will also prepare a presentation that you will present to faculty and students.

Achieved Seminars


2022-09-17_Metaverse – What future is it leading to_by_Dr. Qing Tan  

With Facebook changing its company name to Meta, Meta or Metaverse has become a buzzword. What is the metaverse; what are the enabling technologies; and what are the applications? This presentation will provide an overview of the metaverse and some insight into its enabling technologies as well as the related research and development. The presentation will raise the question of what future the metaverse leads to; and facilitate discussion.

2022-10-15_ Knowledge Graph Embedding for Explainable Multimodal Interactions_by_Dr. Xiaokun Zhang

Knowledge graph (KG) is a knowledge base that uses a graph-structured data model or topology to integrate data along with encoding the semantics underlying the used terminology. Specifically, a KG is a multi-relational graph composed of entities (nodes) and relations (edges) in which each edge is represented as a triple of the form (head entity, relation, tail entity) and the entities are connected by the specific relations. One of the main benefits of using KG instead of typical relational settings is the flexibility towards the schema that allows more flexible data maintenance, evolution, and capturing of incomplete knowledge. The KG related concept was already present in the literature since the ‘80s, but the term KG has regained popularity since Google's Knowledge Graph in 2012. Recent years have witnessed rapid growth in knowledge graph (KG) construction and applications, including the use of KG as tools for explainable machine learning. 

Although effective in representing structured data, the underlying symbolic nature of the triples in the KGs usually makes the graphs hard be manipulated, especially when the growing size and changing interrelations of the data are getting significant. To address this issue, knowledge graph embedding (KGE) based approaches have been proposed and quickly gained massive attention in recent years. KGE is an approach to transform KG (nodes, edges, and their feature vectors) into a low dimensional continuous vector space. Motivation behind KGE is to preserve various graph structural info to make it easier manipulate the information. 

This presentation introduces the concepts of KG and KGE, analyses two KGE models (translation-based model and semantic matching model), explains the KGE applications, and addresses the potential research questions on the following topics and application scenario, such as 

• Link Prediction 

• Triple Classification 

• Recommendation Systems 

• Question Answering 

• Explainable Reasoning over Knowledge Graphs 

• Context-Aware Knowledge Graphs Embedding 

• KGE based Approaches to Explainable Human-Computer Interaction, Multimodal Interaction, and Semantic Interaction in Smart AR/VR/Metaverse 

• KGE based Applications for eLearning and Virtual Lab 

2022_11_12_How Technologies Work_Dr. Jon Dron

Most of us in the field of computing use the term ‘technology’ every day and we usually seem to have a rough idea of what we mean. However, when challenged, even experts in the field find it challenging to give a definition that encompasses all the many meanings and uses of the term. It is, as David Nye puts it, “an annoyingly vague abstraction”. In this talk I will present a different way of thinking about technologies and our relationships with them: one in which we are participants in their enaction, rather than simply users of them; one in which not only are technologies seen to be part of us, but we are seen to be a part of them. By better understanding the nature of that participation, we may be more able to create technologies that meet our needs, and less likely to create those that fail to do so.

2022-12-10_H-AI: Hybrid Human-Artificial_by_Dr. Oscar Lin

Artificial intelligence (AI) is envisioned to enhance human intelligence with quick data collection, analysis, and translation into meaningful insights and actions. While AI methods are being applied to numerous areas, AI researchers and practitioners have been facing a broad range of issues from human interaction with AI. Hybrid human-artificial intelligence (H-AI) aims to research and develop intelligent systems that augment rather replace human intelligence. These systems are developed to leverage human strengths and compensate for human weaknesses. H-AI is an evolution of human-computer interaction (HCI) that integrates both human intelligence and AI into a single entity, thus forming a new enhanced intelligence. Yet, research on H-AI is still in a very early stage, and there are numerous challenges in conceptualizing and developing such systems. In this talk, I will introduce the concept of hybrid human-artificial intelligence and the research opportunities and challenges and discuss the use of the future of AI using intelligent educational systems as an example.  

2023_01_14_Machine Learning Applications in Distance Education_by_Dr. Ali Dewan

Three popular AI applications in education are educational dashboard design, students’ dropout prediction, and student engagement detection. An educational dashboard is used to display educational data in a useful way which allows teachers and students to monitor their online teaching and learning behavior patterns. Student’s dropout prediction aims to predict the student dropping tendency in an online course whereas the student’s engagement detection systems detect the engagement level of the students using different modalities. This presentation will discuss a few of the ways that machine learning approaches are being used in these applications to support online learning along with their challenges and future opportunities. Two major issues - the explainability of AI and the biasness of data in ML techniques – will also be discussed with respect to the above applications.

2023-02-11_Understanding deep learning_by_Dr. Dunwei (Grant) Wen

In this presentation, we will discuss the key mechanisms of deep learning from a technical perspective and explain why they help make a big difference in machine learning and AI. We will then discuss some recent progress in deep learning algorithms, architectures, paradigms, and their impact through several breakthrough applications. We will also discuss the state-of-the-art deep learning tools and applications, and look at the challenges and future research directions for advancing deep learning as well as AI in general.

2023_03_11_Zoned role-based system design and access control_by_Dr. Harris Wang

Nowadays all businesses heavily depend on Web-based systems. For big organizations like universities, governments, and corporations there are many departments with different functionalities and responsibilities. It is often that these departments interconnect with each other when fulfilling certain tasks. As such, the web-based systems for such organizations can be difficult to integrate. When several different systems are implemented, authentication and access control can be very difficult. It becomes even worse when one user has roles in different departments. This is a proven case at this university.

In this talk, I will present a new approach to easing the design and implementation of integrated web-based systems for big organizations. The new approach is called zoned role-based system design. In this approach, departments or functional units are called zones, and for each zone, a set of roles are defined, web apps are designed and implemented for each zone to conduct their business, and each app is associated with some given roles. Each user within the organization is affiliated with a specific zone or zones, with designated roles. As a result, with only one authentication a user will be able to access the necessary apps for each designated role in each affiliated zone.

2023_04_01_Social Engineer Effect: Let’s go Phishing_by_Mr. Ionut Anghelache  

Join us for a jaw-dropping, eye-opening live demonstration of a Social Engineer attack that will leave you shocked and amazed! This presentation will be showcasing cutting-edge cyber security tools like SET, Zphisher, Blackeye, MaskPhish, and Trape to illustrate how easy it is to deceive unsuspecting victims with common internet tricks like phishing and vishing. With the help of our state-of-the-art Credential Harvester, the presentation will redirect users to malicious sites and capture their sensitive information, including their geotag location with pinpoint accuracy. Don't miss this great opportunity to witness the power of Social Engineering in action and learn how to protect yourself from these devastating attacks.

2023-05-13_Personalization and Adaptivity in Learning Systems_by_Dr. Sabine Graf

Personalization and adaptivity allow systems to provide users a personalized experience by first identifying user characteristics and needs, and then using this information to provide personalized pathways, individual recommendations, or adaptive interfaces. Such systems can consider a variety of different user characteristics and needs (e.g., knowledge, goals, motivation, learning preferences, etc.), and there are many different ways in which such systems can be personalized. In this talk, we will look into how such adaptive and personalized systems work and how they can be developed. We will focus on adaptive and personalized systems in the educational domain and demonstrate some of such systems. By doing so, we will discuss how they work and how effective they are in supporting learning. We will also talk about challenges in this area as well as future research opportunities.

2023-06-10_The Science of Crime: Forensic Investigation in the Digital Age_by_Mr. Ionut Anghelache

In this Seminar, we will explore steghide and stegosuite to reveal concealed messages within images, employing the technique of steganography. Additionally, we will utilize Foremost to recover lost data from a disk drive in the .dd format. Autopsy will be employed to recover data and identify hidden parameters, with the assistance of Lime and DD in creating the image. To examine the contents stored in the RAM, we will employ Volatility and SANS. Lastly, for reliable recovery of pictures and videos, we will make use of Bulk Extractor and PhotoRec. We will also engage in a forensic exercise. The objective is to attempt to breach a Virtual Machine and uncover any hidden clues present within it. This may involve searching for hidden comments, concealed pictures, and audio files, and performing data acquisition techniques.

2023-09-09_Can explainable AI offer remedial feedback to learners in academic writing? Quick answer - not yet!_by_Dr. Vive Kumar

Automated writing evaluation (AWE) is a complex research topic that has a long way to go before becoming a mature and trustworthy technology. Research focuses on reporting the performance of AWE systems by measuring their predictive accuracy, without really analyzing what the algorithm actually learned. This lack of transparency is partly due to the high complexity and dimensionality of AWE and the usage of deep learning to cope with them just exacerbated the problem. Interestingly, the ascent of explainable artificial intelligence (xAI) allows us to retrospectively look at the way AWE is developed. However, xAI is itself in its infancy facing many limitations. Consequently, it is important to understand these limitations and how to correctly interpret black-box scoring models. Keeping this balancing act in mind, this presentation introduces key ideas on xAI offering remedial feedback to students. It then ventures into evaluating the utility of xAI in such contexts. 
 

2023-10-14_Using Machine Learning Tools in the Cloud: Experience Gained from the Ask4Summary research project_by_Dr. Maiga Chang

Ask4Summary (https://ask4summary.vipresearch.ca/) is a research that has a system periodically running backend services to process text-based content (e.g., a course's learning materials and the CORD-19 dataset). The CORD-19 dataset includes a growing number of academic articles regarding Coronaviruses; at present, there are more than 717,000 full-text articles in the CORD-19 dataset. When Ask4Summary processes this text-based content, it uses Natural Language Processing (NLP) techniques that include tokenization, n-grams extraction, and part-of-speech (PoS) tagging. It then identifies the keywords from a user’s question and uses cosine similarity to summarize the associated content and present it to the user. When the Ask4Summary is fed with course materials like PDFs, Word, and PowerPoint documents, it can provide summaries for students' questions related to the course. Currently, Ask4Summary can serve users in three ways: web system, Moodle plugin, and chatbot. In this talk, I would also like to share some of our experiences and insights from using Amazon's Comprehend Keyphrase Extraction and Syntax Analysis APIs in the Ask4Summary research project. 

2023-11-18_AI Assists ECE Research - Build an Intelligent Research Knowledge Database for Yourself_by_Dr. Yuzhuo Li

In the rapidly evolving landscape of information technology and artificial intelligence (AI), the need for efficient, personalized knowledge management systems is more pressing than ever. This talk aims to introduce students to the transformative potential of AI-assisted research databases. Guided by real-world demos, attendees will explore how AI tools can facilitate a more streamlined, efficient, and insightful research process. The discussion will extend to the concept of DIY AI-based databases, empowering students to build their own intelligent research knowledge bases without the need for coding expertise. By the end of this session, students will not only understand the utility of AI in research but also feel encouraged to leverage these tools for their own academic and professional development.

2023-12-09_Intrusion Detection and Prevention in Industrial Internet of Things_by_Mr. Nicholas Jeffrey

The Industrial Internet of Things (IIoT) brings the ubiquitous connectivity of the Internet of Things (IoT) to industrial processes, optimizing manufacturing and civil infrastructures with assorted "smart" technologies.  This ubiquitous connectivity to industrial processes has increased, which opens the door for potential cyber threats, making critical infrastructure more susceptible to attacks. The difference between IoT and IIoT is largely one of degree, with the consequence of service interruptions to IoT (ie home automation) typically limited to mild inconvenience, while interruptions to IIoT environments (ie power grids) have more significant economic and life safety consequences. The field of Intrusion Detection Systems / Intrusion Prevention Systems (IDS/IPS) has traditionally focused on cyber components rather than physical components, which has resulted in threat detection capabilities in IIoT environments lagging behind their non-industrial counterparts, leading to increasingly frequent attacks by threat actors against critical infrastructure.  This discussion will review the current state of IDS/IPS capabilities in industrial environments and compare the maturity and effectiveness of the more established IDS/IPS capabilities of non-industrial Information Technology (IT) networks. 

2024-01-13_Enhancing Application Agility: Mastering Feature Management with Azure App Configuration_by_Mr. Oussama Zeaiter

In the rapidly evolving landscape of software development, the ability to manage and control feature rollouts is crucial for maintaining agility and responsiveness. This presentation delves into the world of Feature Management using Azure App Configuration, a pivotal tool in modern application development. We will explore the best practices for implementing feature flags, their impact on continuous integration and deployment (CI/CD) pipelines and the role they play in a DevOps environment. Case studies highlighting successful real-world applications will provide practical insights into the challenges and solutions in feature management. The discussion will also cover security considerations, scaling strategies, and the lifecycle management of feature flags. Additionally, we'll look ahead at the future of feature management, particularly the integration of automation and AI to enhance decision-making processes. By the end of this session, attendees will gain a comprehensive understanding of how Azure App Configuration can be leveraged to enhance application development, improve user experiences, and drive positive business outcomes in a cloud-centric world.

2024-02-10_Controls on AI development_by_Dr. Stella George

It’s fair to say that governance around AI development and AI’s use is massively way behind the changes we have seen in AI development and uptake in the last 7 years. With the recent approval of the EU act and the publication of the US’ presidential order,  the landscape of AI policy is becoming more defined and the controls placed on AI development are beginning to emerge. But, how will these mechanisms of governance impact those who develop AI (both engineers and tech organizations) and end users of AI?  Who will eventually call the shots on AI controls?

20240309_Navigating Cybersecurity: Challenges, Skills, and Future Strategies_by_Mr. Robert Kemp

In today's tech-centric world, having strong cybersecurity practices is a must to protect sensitive information and to keep organizations safe. This talk covers what is happening in cybersecurity, the challenges, and the skills and tools needed to tackle ever-evolving cyber threats. The talk will review why cyber-attacks are on the rise and how to adapt to the ever-changing cybersecurity landscape. How are large organizations reacting to the change needed to deal with setting up strong cybersecurity teams? With cyber incidents increasing rapidly, we will explore new approaches such as Artificial Intelligence(AI), but also look at the challenges that come with it. But will it be enough to stop the tide? In summary, this presentation aims to enhance our understanding of cybersecurity challenges, explore innovative approaches, and highlight possible ways to respond to the evolving landscape.