Nor Aniza Abdullah

Building an E-learning Recommender System Using Vector Space Model and Good Learners Average Rating

An enormous amount of learning materials in e-learning has led to the difficulty on locating suitable learning materials for a particular learning topic, creating the need for content recommendation tools within learning context. In this paper, we aim to address this need by proposing a novel framework for an e-learning recommender system. Our proposed framework works on the idea of recommending learning materials based on the similarity of content items (using Vector Space Model) and good learners’ average rating strategy. This paper presents the overall architecture of the proposed system and its potential implementation via a prototype design.

Is simple sequencing simple adaptive hypermedia

In this paper, we explore the differences between the Adaptive Hypermedia and IMS Simple Sequencing approaches. Both approaches provide learning material tailored for the learners current context. Understanding the difference between the approaches enables us to identify the best features of each, and thus to identify research agendas for the improvement of adaptive hypermedia and of Web-based Learning Management Systems.

Personalisation of MOOCs

Researchers in the field of educational technology are paying huge attention to the widespread adoption of Massive Open Online Courses (MOOCs) in the study of learning online. MOOCs are discussed in many angles including pedagogy, learning sustainability, and business model. However, there are very few discussions around MOOCs personalisation. In this paper, it is aimed to examine and analyse the literature on personalisation of MOOCs to identify the needs, the current states and efforts to personalise learning in MOOCs. The findings denote that the pedagogical design of MOOCs is currently insufficient due to massive and geographically dispersed learners with diverse educational backgrounds, learning requirements and motivations. Many believe that personalisation could address this lacking in MOOCs. Among the most popular services being proposed or implemented in the literature are personalised learning path, personalised assessment and feedback, personalised forum thread and recommendation service for related learning materials or learning tasks.

Augmenting SCORM manifests with adaptive links

This paper describes an experiment to augment SCORM manifests with adaptive links using AuldLinky in order to promote content reusability, interoperability and personalized e-learning. Our technique involves the automatic deduction of a concept map from a manifest and the transformation of its pertinent elements into FOHM (Fundamental Object Hypermedia Model) objects before augmenting the information with complimentary and adaptive links using AuldLinky.

A Toolkit Design Framework for Authoring Multimedia Game-Oriented Educational Content

The learning content such as educational games has the power of engaging students as it allows them to explore and try out things that they have learned in a save and friendly environment. However, this kind of interactive content is difficult to build. Even though there are some authoring tools available, their development environments are still complicated for someone without any basic computing background. A much simpler to use toolkit is therefore urgently needed. We aim to address this need by designing an authoring tool that can operate as simple as Microsoft Power Point, in order to help teachers to produce and personalize their own game-oriented educational content. In this paper, we propose a framework in designing such toolkit. Based on this framework, a toolkit will be developed to enable teachers in Malaysia to author educational game content for teaching basic Jawi words for young.

How Learners’ Interactions Sustain Engagement: A MOOC Case Study

In 2015, 35 million learners participated online in 4,200 MOOCs organized by over 500 universities. Learning designers orchestrate MOOC content to engage learners at scale and retain interest by carefully mixing videos, lectures, readings, quizzes, and discussions. Universally, far fewer people actually participate in MOOCs than originally sign up with a steady attrition as courses progress. Studies have correlated social engagement to completion rates. The FutureLearn MOOC platform specifically provides opportunities to share opinions and to reflect by posting comments, replying, or following discussion threads. This paper investigates learners’ social behaviors in MOOCs and the impact of engagement on course completion. A preliminary study suggested that dropout rates will be lower when learners engage in repeated and frequent social interactions. We subsequently reviewed the literature of prediction models and applied social network analysis techniques to characterize participants’ online interactions examining implications for participant achievements. We analyzed discussions in an eight week FutureLearn MOOC, with 9,855 enrolled learners. Findings indicate that if a learner starts following someone, the probability of their finishing the course is increased; if learners also interact with those they follow, they are highly likely to complete, both important factors to add to the prediction of completion model.

Content-Based Image Retrieval Based on Electromagnetism-Like Mechanism

Recently, many researchers in the field of automatic content-based image retrieval have devoted a remarkable amount of research looking for methods to retrieve the best relevant images to the query image. This paper presents a novel algorithm for increasing the precision in content-based image retrieval based on electromagnetism optimization technique. The electromagnetism optimization is a nature-inspired technique that follows the collective attraction-repulsion mechanism by considering each image as an electrical charge. The algorithm is composed of two phases: fitness function measurement and electromagnetism optimization technique. It is implemented on a database with 8,000 images spread across 80 classes with 100 images in each class. Eight thousand queries are fired on the database, and the overall average precision is computed. Experimental results of the proposed approach have shown significant improvement in the retrieval performance in regard to precision.

Performance evaluation of the combination of Compacted Dither Pattern Codes with Bhattacharyya classifier in video visual concept depiction

High dimensionality and multi-feature combinations can have negative effect on visual concept classification. In our research, we formulated a new compacted form which is Compacted Dither Pattern Code (CDPC) as a chromatic syntactic feature for visual feature extraction. The effectiveness of CDPC with Bhattacharyya classifier for irregular shapes based visual concepts depiction is reported in this paper. The proposed technique can reduce feature space and computational complexity while maintaining visual data mining and retrieval accuracy in high standard. Our system was empowered with Bhattacharyya classifier which has improved efficiency by considering one numeric value which is the Bhattacharyya coefficient. Experiments were conducted on various combinations and compared with different visual descriptors and classifiers. The first experiment illustrates the comparison of the CDPC based results with well known feature space reduction classes. The second and third experiments demonstrate the effectiveness of our approach with multiple perspectives of performance measures including various concepts.

Jawi Character Speech-to-Text Engine Using Linear Predictive and Neural Network for Effective Reading

Jawi is an old version of Malay Language Writing that need to be preserved. Therefore, it is important to develop tools for teaching kids about Jawi characters and Speech-To-Text (STT) application can serve this purpose well. Unlike English, Jawi uses special characters similar to Arabic Characters. However, its pronunciations are in Malay Language. This uniqueness makes STT development a challenging task. In this paper, we investigate the applicability of Linear Predictive Coding to extract important features from voice signal and Neural Network with Backpropagation to classify and recognize spoken words into Jawi Characters. A total of 225 samples of words in Jawi Characters are recorded from speakers with over 95% accuracy. Jawi Characters Speech-To-Text Engine aims to help students to read Jawi document accurately and independently without the need for close monitoring from parents or teachers.

An authoring toolkit design for educational game content

A new way of teaching using multimedia presentations or game-based technologies are urgently needed in order to engage and sustain students’ learning interests in their learning materials. To achieve that, teachers should be able to produce interactive multimedia learning materials (e.g. educational games) themselves. However, current authoring tools such as Adobe Flash and Macromedia Director are too complex to use by teachers with no computing backgrounds. We would like to address this issue by proposing a simple to use authoring toolkit that will enable teachers to create their own educational game contents. The proposed toolkit operates by using games templates i.e. akin to producing presentation slides using Microsoft Power Point. In this paper, we present the framework in designing such toolkit. Based on this framework, an authoring tool will be developed to enable teachers in Malaysia to author educational game content for teaching basic Jawi words for the primary school children.