Deaf students apparently experience hardship in conventional learning; however, despite their inability to hear, nothing can stop them from reading. Although they perform impressively in memorizing the information, their literacy and reading capability still appear to be weak since they lack the chance to revise by listening and practicing repetitively. Currently, the teaching media for deaf students are quite rare and inadequate, forcing them to face difficulties in integrating new knowledge, even though most of the contents are in a form of written, printed, downloaded, or even accessible via an e-learning platform. However, it is crucial to bear in mind that each learner is different. There is evidence showing that some learners prefer particular methods of learning, also known as learning preferences or learning styles. Thus, the present study reports the sequence of learning styles obtained by using a modified VRK + TSL model that categorized students based on their learning styles. We also propose four different ways of teaching using content-adaptive learning styles, namely visual, reading/writing, kinesthetic, and Thai sign language. Based on personal preferences and the principle of universal design under synthesized learning, an e-learning model was developed to identify deaf learners’ learning styles. The objective is to provide e-learning to identify the learning styles of hearing-impaired students and to respond with up-to-date e-learning materials that can be used anywhere and at any time. These materials must support the education of deaf students. As a result, learners have increased efficiency and increased learning outcomes.
This study presents the development of a novel integrated data fusion and assimilation technique to classify learning experiences and patterns among deaf students using Fleming’s model together with Thai Sign Language. Data were collected from students with hearing disabilities (Grades 7–9) studying at special schools in Khon Kaen and Udon Thani, Thailand. This research used six classification algorithms with data being resynthesized and improved via the application of feature selection, and the imbalanced data corrected using the synthetic minority oversampling technique. The collection of data from deaf students was evaluated using a 10-fold validation. This revealed that the multi-layer perceptron algorithm yields the highest accuracy. These research results are intended for application in further studies involving imbalanced data problems.