View source for Integrated Telegram and Web-based Forum with Automatic Assessment of Questions and Answers for Collaborative Learning


== Summary ==
<p align="justify">At SMU-SIS, a few of us have developed and implemented an integrated Telegram and Web-based Q&A Forum, capable of automatic thoughtfulness assessment of posts, using text mining and Natural Language Processing methodologies, coupled with intelligent engagement features, to facilitate student-driven participatory and collaborative learning. The system was developed for the undergraduate course COR1305 Spreadsheets Modeling and Analytics (formerly known as IS102 Computer as an Analysis Tool, or affectionately known as the CAT course). Thus, we have tentatively called the system CAT-IT, due to the long history of the CAT course.</p>

== Motivation ==
<p align="justify">One key 21st century skill that students need to have is “the ability to ask and answer important questions, to critically review what others say about a subject, to pose and solve problems, to communicate and work with others in learning, and to create new knowledge and innovations” [1]. One way to develop such skills is through participatory learning which involves “cooperative learning tasks which are dialogic in nature and involve student-to-student interaction, rather than just teacher-student interaction.” [2].</p>

<p align="justify">Asynchronous online discussion forums have been used in many learning contexts to enable students to learn from their peers as a community, and it was mentioned in [3] that “perhaps the most important form of active learning is discussion.” Despite the popularity of using threaded forums in Learning Management Systems (LMS), it was argued that they “might not be the best technology to support the interactive and collaborative processes essential to a conversational model of learning” [4]. 
</p>

<p align="justify">Several suggestions that future discussion environments should improve upon was provided by [5]. 
</p>
<li>First, it is to foster an online community, provide timely feedback, encourage information sharing and support collaborative problem solving. To achieve these objectives, some incentive mechanism may be designed into the environment. </li>
<li>Second, the lack of convergence requires a system which can go beyond knowledge sharing to include active processing and synthesizing of the information provided by the community. </li>
<li>Third, the use of multi-functional environments or systems that can integrate new media technology to facilitate learning at different phases, such as asynchronous and synchronous discussions. </li>

<p align="justify">We target to overcome these ineffectiveness and facilitate active and meaningful online discussions for students at our university, assisting them to acquire this important skill and achieve better learning outcomes. 
</p>

<li>For the first feature, we have purposely designed six in-built incentive and engagement mechanisms in our Q&A platform. </li>
<li>For the second feature, our platform is capable of automatic thoughtfulness (quality measure) assessment of question and answer posts, using text mining and Natural Language Processing (NLP) methodologies, and prompt the students to improve on their posts, thereby invoking deeper thinking and learning, beyond just knowledge sharing.</li>
<li>For the third feature, our platform is an integrated Telegram (a mobile app) and web-based forum which are synchronized real-time, so that student learning and sharing can occur anytime and anywhere.</li>

== Student Engagement Features ==
<p align="justify">The platform, CAT-IT, is a student-centric platform where the interactions in the form of asking questions and providing answers, are all student-driven. We have chosen this design as we believe that while the traditional tutor-student delivery is important, student-student interaction is also important for academic engagement as all participants are equals and there will be no power relationship issues which may hinder active discussions [6]. We have designed six special features to improve and maintain student engagement:</p>
[[File:Figure0.png|center|600x300 px]]

== System Architecture and UI ==
[[File:Figure1.png|framed|center|Figure 1. Schematic Diagram of System Architecture]]
<br><br>
[[File:Figure2.png|framed|center|200px|Figure 2. Web-based Forum to View, Post and Search for Posts, and to View User Info and Ranking]]
<br><br>
[[File:Figure3.png|frame|center|200px|Figure 3. Telegram App Interface for Posting Question and Answering Question]]
<br><br>

== Data Preparation Tasks ==

<li>Data sampling: temporary solution for computational challenge</li>
<li>Data cleaning: data types, demographic fields clean up</li>
<li>Data joining: merge data from different sources</li>
<li>Data transformation: matrix for chord diagram, path for sunburst diagram, transaction table for association analysis, node and edge table for network visualization</li>

== System Process Flow ==
[[File:Figure4.png|frame|center|200px|Figure 4. Process Flow]]
<br><br>

== Publications ==
<p align="justify">We have published the details of our platform and results of our implementation runs in the following papers.</p>
<li>Cheong, M.L.F., Chen, J.Y-C., & Dai, B.T. (2018). Integrated Telegram and web-based forum with automatic assessment of questions and answers for collaborative learning. In: Proceedings of IEEE International Conference on Teaching, Assessment and Learning for Engineering TALE 2018: Wollongong, Australia.</li>
<li>Cheong, M.L.F., Chen, J.Y-C., & Dai, B.T. (2019). An Intelligent Platform with Automatic Assessment and Engagement Features for Active Online Discussions. Presented at IEA/AIE 2019, and published in Advances and Trends in Artificial Intelligence, LNAI 11606, 730-743.</li>

<p align="justify">We are working on more papers to further share our research and outputs.</p>
<li>Cheong, M.L.F., Chen, J.Y-C., & Dai, B.T. Analysis of Online Posts to Discover Student Learning Challenges and Inform Curriculum Improvement Actions – an Alternative Approach.</li>
<li>Dai, B.T., Chen, J.Y-C., Cheong, M.L.F. Insights from Features Evaluation for Posts Improvement through Automated Guidance in a Discussion Forum.</li>
<li>Cheong, M.L.F., Singh, A.V., Chen, J.Y-C., & Dai, B.T. Assessing the Impact of Intelligent Q&A Forum on Student Collaborative Learning.</li>

== Expression of Interest ==
<p align="justify">We welcome other Institutions of Higher Learning (IHL) and organizations to explore how our platform can be adapted for implementation in their environment. Please contact Associate Professor Michelle CHEONG (michcheong@smu.edu.sg) for further discussions.</p>

== References ==
[1] Trilling, B., & Fadel, C. (2009). 21st Century Skills: Learning for Life in Our Times. San Francisco, CA: John Wiley & Sons.<br>
[2] Cahill, H., Coffey, J., Lester, L., Midford, R., Ramsden, R., & Venning, L.  (2014). Influences on Teachers’ Use of Participatory Learning Strategies in Health Education Classes. In: Health Education Journal, 73(6), 702-713.<br>
[3] Kember, D., & McNaught, C. (2007). Enhancing university teaching: lessons from research into award winning teachers. In: Abingdon, Oxforshire: Routledge, UK, C.<br>
[4] Thomas, M.J.W. (2002). Learning with incoherent structures: the space of online discussion forums. In: Journal of Computer Assisted Learning, 18, 351-366.<br>
[5] Gao, F., Zhang, T., & Franklin, T. (2013). Designing asynchronous online discussion environments: recent progress and possible future directions. In: British Journal of Educational Technology, 44(3), 469–483.<br>
[6] Kassens-Noor, R. (2012). Twitter as a teaching practice to enhance active and informal learning in higher education: The case of sustainable tweets. In: Active Learning in Higher Education, 13, 9-21.