Research / Researchers / Publications

Permanent URI for this community

http://hdl.handle.net/11071/2474

Browse

Browsing Research / Researchers / Publications by Author "Abdalla, Radwan"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Improving learning throughput in e-learning using interactive-cognitive based assessment
    (The Society of Digital Information and Wireless Communications (SDIWC), 2014) Ballera, Melvin; Ateya, Ismail Lukandu; Abdalla, Radwan
    An e-learning website is not sufficient to fully attain the results of online education. There also is a need to align the educational objectives into the design of the assessment to improve and develop cognition, critical thinking and problem-solving skills. Previous studies have explored the potentials of the assessment models but few ventured into their implementation. Others only proposed and introduced conceptual frameworks. The implementation of these proposals, however, revealed that the question type in the assessment phase neglected to align their questionnaire formats into a cognitive schema. At present, the standard multiple-choice question is the most frequently used of the question type of e-learning assessments. However, if this type is the only format adopted by e-learning developers, then the potentially rich and embedded assessment of the computer platform will be given up. This paper focuses on the design of assessment questions, which is created and guided by the hierarchical Bloom cognitive taxonomy and by utilizing rich media formats. Results conducted for eighteen weeks show a dramatic increase in the academic performance of the students. Likewise, digital transcripts converted from the collected perceptions after training undergoes sentiment analysis have correlated with the student improved academic throughput.
  • Thumbnail Image
    Item
    Personalizing E-learning curriculum using: reversed roulette wheel selection algorithm
    (IEEE, 2014-09) Ballera, Melvin; Lukandu, Ismail A.; Abdalla, Radwan
    E-learning poses a challenge in a pedagogical perspective such as finding ways on how to motivate the students to learn in spite of the absence of a human instructor. Many researchers in the field have proposed and implemented various mechanisms to improve the learning process such as individualization and personalization. The main objectives is to maximize learning by dynamically selecting the closest teaching operation in order to achieve the learning goals. In this paper, a revolutionary technique has been proposed and implemented to perform individualization and personalization using reversed roulette wheel selection algorithm that runs at O(n). The technique is simpler to implement and is algorithmically less expensive compared to other revolutionary algorithms since it collects the dynamic real time performance such as examinations, reviews and study matrices. Results show that the implemented system is capable of recommending new learning sequences that lessens time of study based on their prior knowledge and real performance matrix.
  • Thumbnail Image
    Item
    Reversed Roulette Wheel Selection Algorithms (RWSA) and Reinforcement Learning (RL) for personalizing and improving e-Learning system: the case study and its implementation
    (The Society of Digital Information and Wireless Communications (SDIWC), 2015) Ballera, Melvin; Lukandu, Ismail A.; Abdalla, Radwan
    Various mechanisms to improve learning process with the objective of maximizing learning and dynamically selecting the best teaching operation to achieve learning goals have been done in the field of personalized learning. However, instructional strategists have failed to address the necessary corrective measures to remediate immediately learning difficulties. It is necessary that an alternative, more realistic, simpler and a real time multi-based performance for personalized learning sequence be developed and implemented. Three major contributions can be asserted by the study: it personalized the learning sequence using reversed roulette wheel selection algorithm and linear ranking; the fitness value is based on real time, multi-based performance system; and it implements the reinforcement and mastery learning to motivate students and to improve their learning output. Result shows that the personalized learning sequence (PLS) were dynamic and heuristic and considers the curriculum difficulty level and the curriculum continuity of successive curriculum while producing individualized and personalized learning sequence. Data collected during 18 weeks experimental sessions, from 34%, a 54% increased has been achieved, making the overall passing rate to 88%. Digital transcripts based on students’ perceptions and experiences in using the prototype positively correlates with theme analysis having a score of +.321 with positive attitude such as: very happy, friends, motivate, improve, understanding, knowledge and good were extracted from document analysis.