Meeting the Minds
Dr. Maiga Chang, Part II
Volume 24 Issue 16 2016-04-22
Athabasca University. (2016). Maiga Chang.
Dr. Maiga Chang is an associate professor in AU’s School of Information and Computing Sciences and is the New Initiative Chair on executive board of the IEEE Technical Committee of Learning Technology. His research interests include mobile learning and ubiquitous learning, museum e-learning, game-based learning, educational robots, learning behaviour analysis, data mining, intelligent agent technology, computational intelligence in e-learning, and mobile healthcare. He recently took some time to be interviewed for The Voice Magazine, and this is the second part of that three-part interview.
How do your research interests integrate into development of courses?
I have one research direction, which is to make computers, not smarter, but to be capable of having emotions and recognizing users’ emotions. Researchers in the United States have done some research that is pertinent. They have tried to develop a computer an assistant. It is quite good. You can ask any question about the course, and the computer can answer you. They also developed another computer, a virtual person. He or she does not know anything. But anytime you have a question, he or she will show his or her emotion. You say, "This part is so difficult. How will I learn?" He or she will say, "Oh, you can definitely find an answer on your own. Why not take a look at your textbook once again?" You say, "I think I figured this out." He or she says, "Oh my god, I am so glad to see that." So, it is quite intelligent, but it actually doesn’t know anything.
The researchers want the students to feel that they are interacting with a teaching assistant and never tell them that in fact it is a computer they are interacting with. At the end, the researchers ask the students, "Tell us the name of the smartest teaching assistant you think" The students say the one that has emotion. This is important. This is why we have this kind of research. We have to make sure the computer has emotion. Also, they need to identify your emotion. From your typing, from your voice, from your facial expression, this kind of research I created a course. It is Affective Computing. If you take that course, at the end of the course, you should be able to develop computer guy with emotion detection, function, or has its own emotion.
Also, as you know, my research topic includes mobile application, personal service, location-based service. So, I created a course called Mobile Computing, and another called Mobile Game Development. Based on my research, I’ve created these kind of courses. So, students get some ideas, or get engaged, saying, "Oh! Now, I have a computer, which has emotion. Can I do more?" They will head off into another level of research to start working on.
What are the benefits of teaching at AU compared to traditional universities?
There are differences. They are different from traditional university and AU because we are almost purely online as a university. We teach students with a lot of help from technology. So, in that case, I would say that teaching at AU that we are the pioneers of teaching students with technology, artificial intelligence applications, learning analytics – everything. I would say that this kind of teaching and learning should be the future. As you know, some people start to work on full time jobs after K-12 and some of them go to university for another four years, which means they only learn in traditional classroom or in traditional setting for 12 to 16, maybe 18 years.
How long will you live? How long will you need to learn? You will need to learn for your whole life. When you graduate from high school and university, you cannot go back to university unless you want to quit a job when you want to learn once again. You will need another way of doing life-long learning.
AU gives us the opportunity to create a kind of smart learning environment. So if we can use our research results to make a smarter learning environment, then we can provide students with more personalized learning experiences, which can make them learn more efficient, and learn the things that they really need and want to see on their own way and own pace. That is another good thing for students, I would say, teaching at AU.
What do you think are the strengths of learning at AU?
This is the future. Like the students right now in high school and in primary school, you can ask them. They are trying to use mobile devices to learn. Also, as you know, they will post something on their Facebook or their blog. That is the future. As a parent, around 50% of students at AU have family, even children. When they learn at AU, they are adapting to the future of learning, and, in that case, when their child or children have a question. In my upbringing, I could not ask questions of my parents about using Facebook, but right now, you can, because people use Facebook. Now when you’re taking an AU course, you are sometimes asked to make a video, put it on YouTube, and then you can teach your children, your child.
One more thing is very important. It is self-regulated learning skill. It is very important for everyone because it helps you efficiently learn, or digest, or plan your goal. When you learn with AU, you will learn that kind of skills. You can teach your child and children, and other family members.
You worked in the Machine Intelligence Research Labs (Machine Intelligence Research Labs, 2013). What did this position involve in terms of tasks and research topics?
The lab, actually, is a kind of Special Interest Group (SIG) organization. It is free for any researchers that have shared research interests to join. In this lab, researchers can share their research results. They can share the research and relevant opportunities they heard, and they can work with others if they have a kind of project and they are looking for collaborators.
What is the general process of research that is important for undergraduate students to know?
The process starts with an idea. You will have an idea through seeing something, from newspaper, television or even a movie. Have you seen the television series called Person of Interest? It involves a lot of artificial intelligence, voice recognition, text mining, and data analytics themes. Then, you survey some relevant literature, which makes you more comfortable with your idea, because if you can see research that was done by others, research identical to your idea, then you do not need to do it. When you read the literature, you will have more confidence in being able to do it. Then, you need to find a specific goal for your idea. So, what do you want to do with your research?
When I supervise students, I ask them to think big, but, at the end, when they start doing research they start focusing on the specific idea of that research. When you have a goal, you need to break it down into specific objectives. For instance, if I have a goal, say, I want to have world peace. But to achieve that, I need to break it into three or four objectives such that when I finish each objective, I will be closer to my goal. When all of the objectives have been finished, or achieved at the end, then I should have reached my goal. For any objective, we probably have some kind of issues, or some sort of "problems" needing to be solved. So, for example, I want to create a virtual person that can learn from reading a book. In order to do that, I need to solve some issues. For instance, how does the computer store the text or the image when it read a book? If I can solve all of the issues, then the objective can be accomplished. Of course, once you identify the issues, you need to do more in-depth literature review or surveys. Because sometimes you will see people have solved this particular research issue, but not very well or with limitations, and you can improve this. Or perhaps, you can mix three or four methodologies for better solutions to your issues. You need to do more in-depth or comprehensive surveys after you have identified the issues and possible solutions.
The most important part is, once you finish all of the objective, you need your goal to be evaluated. You cannot simply say, "Okay, I am done! I think this is good!" However, when you do the pilot, when you ask people to really use the system, you will probably find a lot of problems. People will say, "No, this is not user-friendly. It is not understanding me correctly." In that case, you need to evaluate your research outcome such like a system, a tool, or a framework and workflow. When you analyze your system or research results, you need to use appropriate measure. For instance, we have quantitative and qualitative research methodologies. You need to evaluate your system in different ways. For example, if you are doing a searching program, like Google search engine, then you can try to evaluate your system by using the performance. How much time it needs to find something? How accurate is the search result or results? Also, you can do that from a qualitative way. That is basically the process of research. A very difficult one. (Laughs)
For undergraduate students hoping to become engaged in research at AU and beyond, what resources are available through AU for undergraduate students?
AU Library. They provide free access to most of the academic papers and data. If you want to do the research, and you want to engage in the research, you can access all of the academic research work in the world without any problems. Also, in my case, some will come to me and say he or she wants to do research. And I coincidentally have some research projects I’m thinking about. So, I hire them to be the research assistant. And they can do or take part in the research project. A couple even published papers–undergraduates—published papers on whatever they did in the conference. I think these resources are what they are looking for.
We have a role called an advisor in the School of Information and Computing Sciences. You can approach an advisor and tell them you want to do some research, and ask what kind of courses you should take or which professors you should contact. When you get the contact, you can talk to the professor. For example, some students ask me. I tell them, "Okay, you should take these three courses. In the final project course, you can do a research project based on what you learn in these three or four courses." Those are the resources available for undergraduate students.
What about graduate students?
Graduate students can access more. We have a graduate student research fund, which can support graduate students during their research, and they can disseminate in their research in conferences. Also, if you are living in Alberta, you can apply for Profiling Alberta’s Graduate Students fund for attending a conference. If you are not living in Alberta, on the other hand, you can apply for Graduate Level Student Travel Awards. They are similar. One more thing, good thing, for MScIS students is that they can apply for Alberta Innovates – Technology Futures (AI-TF) Graduate Student Scholarships, which is $5,000 per year for two years during their thesis research.
Our family of graduate studies. They always hold a lot of online seminars that teach you how to write a paper, how to prepare your research agenda, how to make your research presentation, and so on. So, this additional resource, which can be accessed by graduate students.
How can undergraduate students improve their research and the resources available for their research?
Talking to a professor or advisor is an important part of this, so you can identify the resources needed to access before you start working on the research idea or topic you are interested in. For example, if you tell me, "I want to do this kind of research." Then I will let you know what kind of courses you might want to take, and then look at these two books, and then, of course, I will talk to you to try and elaborate the research a little bit more because a lot of undergraduate students, their research ideas are more practical, which means they look at something and want to do that. And yes, that can be done, but that is not, really, research but kind of duplicating research. So, in that case, we need to try to figure out what part of your idea can be more research-oriented. That is what they can do: talk to and listen to the advice of the professor and advisor. That is most important.
Retrieved from http://maiga.athabascau.ca/. Machine Intelligence Research Labs. (2013). Maiga Chang.
Retrieved from http://www.mirlabs.net/global/index.php?c=main&a=person&id=558.
Scott Douglas Jacobsen is an AUSU Councillor. He works with various organizations, and runs In-Sight: Independent Interview-Based Journal, and In-Sight Publishing.