Zulikha Jamaludin

Students' Data-Driven Decision Making in HEI: The Explicit Knowledge Involved

Due to increase in the volume of students’ data and the limitations of the available data management tools, higher education institutions (HEIs) are experiencing information overload and constrained decision making process. To attend to this, Information Visualization (InfoVis) is suggested as a befitting tool.However, since InfoVis design must be premised on a pre-design stage that outlines the explicit knowledge to be discovered by the HEIs, so as to actualize a functional and befitting InfoVis framework, this study investigates the explicit knowledge through survey questionnaires administered to 32 HEI decision makers.The result shows that relationship between the students’ performance and their program of study is the most prioritized explicit knowledge, among others.Based on the findings, this study elicits a comprehensive data dimensions (attributes)expected of each data instance in the HEI students’ datasets to achieve an appropriate InfoVis framework that will support the discovery of the explicit knowledge.Our future work therefore include designing the appropriate visualization, interaction and visual data mining techniques that will support the explicit knowledge discovery and HEI students’ data-driven decision making types.

Improving ASR performance using context‐dependent phoneme models

Purpose – The purpose of this paper is to present evidence of the need to have a carefully designed lexical model for speech recognition for dyslexic children reading in Bahasa Melayu (BM).Design/methodology/approach – Data collection is performed to obtain the most frequent reading error patterns and the reading recordings. Design and development of the lexical model considers the errors for better recognition accuracy.Findings – It is found that the recognition accuracy is increased to 75 percent when using context‐dependent (CD) phoneme model and phoneme refinement rule. Comparison between context‐independent phoneme models and CD phoneme model is also presented.Research limitations/implications – The most frequent errors recognized and obtained from data collection and analysis illustrate and support that phonological deficit is the major factor for reading disabilities in dyslexics.Practical implications – This paper provides the first step towards materializing an automated speech recognition (ASR)‐...

Structural-Semantic Analysis: A Method Toward a Classification of Malaysian Folktale

This paper expounds a method that will be used in a study to propose a Malaysian folktale classification system focusing on Malay folktales. At core, there are three significant folktale classification systems that classify based on three different folktale units: type, motif, and function. Individually, type and motif each classifies based on the content of the folktale, whereas function, the structure of folktale. The study intends to generate the folktale classification system for the Malay folktales based on an integration of the three distinguished units of the folktale classification system. In order to achieve that, the method opted is structural-semantic analysis. The method, which constitutes of three levels of classification, classifies sequentially according to the content and the structure of the Malay folktales. It is hoped that the method adapted would engender a holistic and robust Malaysian folktale classification system.

Embedding interaction design in Wayang Kulit mathematics courseware

Digital technology environment has emerged as a great structure for digital learning and teaching (L&T) mathematics with the support of multimedia interaction. Wayang Kulit (WK) is unlikely to last long due to lack of safeguarding. The traditional WK has a great potential to be incorporated into a digital L&T Mathematics (dLTM) in primary school. This is intended to create a fun and appealing method in the process of learning mathematics. Henceforward, we proposed a conceptual framework that illustrates how digital WK helps students learn mathematics. We explored and developed a digital WK by integrating the core elements of WK into L&T mathematics environment. In order to support the engagement and entertainment in dLTM, the needs for effective interaction design with five attributes of Interaction Design (IxD) theory are added. Afterwards, the survey questions based on Usefulness, Satisfaction, and Ease of Use (USE) were distributed to measure the performance of dLTM. A framework that mapped WK in a dLTM environment, including the useful (79%) and satisfying (83.4%) consequences for such undertaking and endeavor, was also conferred. The result showed that the students perceived high satisfaction towards the proposed digital WK.

Modelling a digital graphic novel

Researchers and designers proposed “digital graphic novels” (DGN) because reading a graphic novel from physical book is not a popular choice anymore due to the absence of interactivity, user control and rich multimedia support. It however opens ups debates on whether such graphic narratives is usable as well as effective for transferring knowledge. Researchers in Human-Computer Interaction field argue that this kind of media enables the absorption of previous knowledge in the story presented and can suggest new knowledge to the user. However, not designing DGN based on a solid model will continue to bring prejudice for this media. This paper focuses specifically on cognitive processes of constructing a conceptual DGN model of an explanatory type. Utilising action research cycle, Rigauds unsupervised framework of DGN, Kanes educational graphic narrative model, Hughess dual pathways model, and Cohns visual graphics narrative model were examined and analysed to extract variables that can explain DGN semantics. The analysis revealed that the unsupervised framework can be extracted to improve newer model. We thus enhance Kanes educational graphic narrative model in order to translate the knowledge into a target system. Expert evaluation was carried out on the enhanced model. The evaluation results provide positive insights specifically on idealising assumptions and viability of the model. We then proposed the final premise where visual narrative comprehends the meanings of sequential images characterised by order and timing delivered in the storytelling strategised by graphic narrative and graphic text consisting of images, content and print texts in the knowledge base. This premise of the model serves as one of the principle instruments of modern science, with specific contribution to the development of usable DGNs. The model contains rational attributes to be used as a foundation for DGN design and development.

The design of mobile social application for children with autism

Autism is a neural syndrome that complicates the growth of mind, producing challenging result in communicating, social interaction, and impairment in behaviour. Since there is no cure for autism, prompt interventions and effective educational exercises allow children to achieve massive improvement. During the teaching and learning process, children with autism require particular consideration and attention. Thus, with the implementation of information technology in special education, the teaching and learning process could to become more efficient. Struggle in social interaction skill is known as one of the main drawback encountered by children with autism. Therefore, a mobile social application is proposed to help the children to develop social interaction skills. Nonetheless, this paper only deliberates on the design phase mobile application development. It is developed for children with medium functioning Autism Spectrum Disorder (ASD) who are learning basic interaction skills. The applications design phase implements the results from the analysis phase, which has concluded earlier. Five evaluators were involved in the heuristic evaluation, in order to enhance the user interface of the mobile social application. The outcomes from the evaluation conclude that all evaluators has agreed with all heuristics. Not to mention, beneficial recommendations were also achieved from the evaluation. The future work of this paper will be the development phase of the mobile social application.

An Algorithm to Define Emotions Based on Facial Gestures as Automated Input in Survey Instrument

This study is carried out to help researchers who are having problem in acquiring accurate responses or answers from dyslexic children in their survey instrument.Due to the children impairment in reading, spelling and writing, researchers who conduct studies on this particular group are unable to gather valuable information from their subjects.Therefore, this paper examines on the use of facial gestures as input in survey instrument. Features of facial gestures that play important role in creating emotions are selected to be formulated in the algorithm.The aim of this paper is to produce an algorithm to define emotions based on three facial features namely mouth, eyes, and eyebrows.This is done by studying users and the problem domain, analyzing the relationship between emotions and facial gestures, and formulating solutions to the problem.This algorithm will be useful in developing a more interactive survey instrument and improving user interaction between researchers and dyslexic children.

A conceptual framework for students’ data-driven decision making process in higher education institutions

With the increase in volume and the wide range of data needed to inform decisions at all levels of education system, federal and state ministries of education, districts and also schools have been demanding proactive and data-driven decision making (DDDM) frameworks, as an approach to enhance policy and decision making.In dealing with DDDM, higher education institutions (HEI), among schools, are found to be the most in need of such decision-support frameworks, because of the volume of the data they deal with, and the sensitivities of the decisions they make.To attend to this, though with an exclusive focus on the HEI’s students, we propose a DDDM conceptual framework, using a Design Research Method.The method includes critical literature review, survey instrument administration on the HEI policymakers, and interview sessions with data management personnel. Our proposed conceptual framework shows (1) how DDDM can achieve effective students’ data usage and improved students-related decision making, (2) HEI decision makers’ data usages, their preferred explicit knowledge from the data, and the relationship with the appropriate data variables.

BMTutor research design: Malay sentence parse tree visualization

This paper discusses the research design for BMTutor. BMTutor is a prototype for visualizing Malay sentences that is combined with sentence checker, sentence correction and word attribute components. The purpose of BMTutor is to check sentence validation, provide sentence correction for invalid sentence used and produce parse tree visualization. The research design involved can be divided into four phases; categorizing sentence and produce repository (Phase 1), developing models and algorithms (Phase 2); development of a prototype (Phase 3); and prototype testing (Phase 4). To date, this system is the only one designed with the functions and characteristics as in BMTutor. There are two BM parsers to check the validity of simple BM sentences had been developed. Both parsers performed three phases in research design, namely 1) the collection of sentence or CFG, 2) develop a prototype, and 3) conduct evaluation. The phases involved are the basic method in developing a prototype. As a result of the lack of models and algorithms have been introduced in both parsers, the model and algorithm development phase is introduced in the design of BMTutor. Output from the development process shows that the prototype is able to provide sentence correction for all 15 invalid sentences and can produce parse tree visualizations for all 20 sentences used for prototype testing.

Let's Play with Colours: BacaMAX User Interface for Dyslexic Children

Reading difficulties are synonymous with dyslexic children, whose unique condition is due to dyslexia. One of the major theories of dyslexia is visual deficits which are not caused by the problems with the eyes but with the information processing that took place inside the brain. Hence, to assist dyslexic children reading we propose BacaMAX, a visually stimulating voice replay application designed carefully to ease them reading. The application is a result of years of study on dyslexic children and their colour choices. Bedside experiment with the children was carried out on a systematic background and a foreground colour scheme, specifically on syllables, words, and short sentences. This paper aims to deliberately discuss BacaMAX interface design that started with our beta version to the current improved version. Rapid application development is employed as the methodology to develop prototypes and user acceptance test is employed to test the prototypes on real users. As the result, all versions are presented and discussed.