PhD FAQ

• Supporting national agenda and the needs in cross-platform MOOC monitoring and analysis;
  • Malaysian government thru the Ministry of Education Malaysia in Malaysia Education Blueprint 2015-2025 (Higher Education) and in National Learning Policy (DePAN) for Higher Educational Institutions, has plan and needs for MOOC to be empowered and used accordingly align with national agenda and policy.
  • Not only in Malaysia other countries is also starting to look more seriously on MOOC and align it with national needs, including monitoring MOOCs usage.
    • According to Wu Yan, Head of China Higher Education Department of The Ministry, China government also plans to form a list of 3,000 recommended courses by 2020 (moe.gov.cn) [3]
    • The European MOOC Consortium (EMC) recently launched a Common Microcredential Framework (CMF) with its founding platform partners including FutureLearn, France Université Numérique (FUN), OpenupEd, Miríadax, and EduOpen during EADTU-EU Summit 2019, Brussels [8].
  • MOOC becoming global and backbone of the new modern education ecosystem.

• The main issue address in this research is how MOOC can be monitor and analyzed effectively to measure the MOOC performances on cross-platform.
  • How MOOC (regardless of platforms) semantic data and learning analytics can be used to analyze MOOC performances (student performance, course performance, and platform performance).
• Differences learning style and course instructional design effect towards the learners' experience which eventually affect the learners and course performance itself.
• Different MOOC platform also might have different data model and analytic data design.
• By understand how learners react and use the MOOC content based on the performance data, the instructor can design the content more effectively suitable for the learner's engagement and batter end results of the MOOC learning activities.

There are huge collections and still growing numbers of courses created and offered on MOOC platforms. A same keyword topic search, “Python” from three MOOC platforms, Udemy (10,000 results found), Linkedin (formerly known as Lynda. 3,248 results found) and Coursera (228 results found) is an example in numbers of vast collection similar courses offered.

Due to the massive nature of MOOCs, the number of learning activities (e.g. forum posts, video comments, assessment) is becoming difficult to be tracked by the course learners [9] [10] [11] or observed by the course instructors nor course providers. There is also a possibility for learners to overlook on courses that most suitable for them.

At this point in time, not all MOOC platform or its analytic data is supporting adaptive learning. Although numbers of researchers already start looking into adaptive learning to be implemented in MOOC platforms.

Chinese government for example, already start listing recommends online courses to boost MOOC, where the ministry will conduct inspections on the courses regarding their operation, effects, and updates. Courses that fail to meet the standards will be removed from the list (xinhuanet.com) [5].

This research will response to above-mentioned problems with the design and development of a new semantic data model that is dynamically set and derived from MOOC instructional design and MOOC learning analytic to provide input for improving learning performance, teaching performance and course performance towards personalized and adaptive learning, "MOOC Performance Analyser (MOOC-PA)"

1. What are the parameters for observing MOOC performance based on MOOC instructional designs and MOOC learning analytics?

• To identify the parameters, we need to know how do people interact with MOOCs and this can be achieved by conducting a series of surveys, interviews, and field study at a different scope and level of MOOC users in terms of the MOOC platform used, field of subjects, institutions, and user preferences.
• Initially, five (5) MOOC from five universities in Malaysia were identified to take part in this research as sample data, which are Universiti Teknikal Malaysia Melaka (UTeM), Universiti Putra Malaysia (UPM), Universiti Sains Islam Malaysia (USIM), Universiti Malaysia Terengganu (UMT) and Universiti Kebangsaan Malaysia (UKM) and one (1) MOOC from university in United Kingdom, which is University of Southampton.

2. How to define RDF for MOOC with different learners learning style?

• Based on the study and findings in identifying the parameters for observing MOOC performance, MOOC feature will be categorized based on four (4) category of learning style which is Visual, Auditory, Reading/Writing Preference, and Kinesthetic [15]. MOOC dataset will be analyzed based on the findings of how people interact with MOOCs.

3. What is the framework for MOOC engagement and analysis?

• Based on findings and data analysis from previous activities, a framework for MOOC engagement and analysis will be produced. This framework will embed into the existing RDF.

4. Does the framework suitable with another MOOC?

• A series of testing using the same and the different MOOC platform will be carried to evaluate the designed framework.
• The dataset from other MOOC platforms will be used and should produce results as regardless what platform people use, the MOOCs settings and approach that match with people learning style will affect the end result in either the learning process achieved a certain level of engagement and enhance learners knowledge.

Specific objectives and contribution in this program of research are summarized as follows:

• To analyses factors, contribute to the MOOC performance based on learners learning style, MOOC instructional designs, and MOOC learning analytic.
  • It develops the first hybrid model embedding learning style and instructional design together with the semantic web. Here a machine learning method is used to process all analytic and information available that will also be supporting adaptive learning.

• To define the dynamic MOOC Resource Description Framework (RDF) data model for MOOC-PA based on the proposed parameters using Malaysia and United Kingdom MOOCs as case studies.
  • A novel MOOC-PA RDF will be developed which are suitable for cross MOOC platforms. To solve the cross-platform issues, an optimization function has been proposed to produce the analytic parameters. The measured data is generated using 3 MOOC platforms which are FutureLearn, OpenLearning, and edX.

• To develop the MOOC-PA based on designed models.
  • The first cross platforms MOOC analytics scripts run as WordPress plugins using the hybrid model and adaptive learning approaches

• To evaluate the query results experiments on MOOC performance using Linked Data quality measures.
  • The result which will confirm that the hybrid model MOOC-PA RDF is able to improve the analytic processing and provide MOOC performance analyzer data compared to the traditional basic analytic by each MOOC platform. This established that combining information of learners learning style, the MOOC instructional design and existing MOOC analytics data based on a hybrid model can rapidly reduce

The next and future step is to expand the MOOC-PA availability …

• This is applied research.
• This research was designed to solve specific practical problems and answer certain questions. It limited to MOOC analytics and observatory to find out the solution for the research problems statements.

• This research will contribute to the creation of a MOOC Observatory tool for stakeholders (to be used in learning, teaching, and monitoring)
• The success of this research will give implications towards a new era of MOOC platform where adaptive learning been used along with existing MOOC data and at a cross-platform level. It will help the learner to personalize identify the most suitable course for them and help course instructor to better understand their learners and monitor the course engagement based on the student performance, course performance, and MOOC performance analyzed results.