Gerhana, Yana Aditia (2025) An Artificial Intelligence-Based Knowledge Management System for Outcome-Based Education Implementing in Higher Education Institutions. Doctoral thesis, Asia e University, Malaysia.
![[img]](http://ur.aeu.edu.my/style/images/fileicons/text.png) |
Text
Thesis Phd_Yana Aditia Gerhana (reviewed and edited) 170925 (1).pdf Restricted to Repository staff only Download (9MB) |
|
Text
Thesis Phd_Yana Aditia Gerhana (reviewed and edited) 170925 (1)-1-24.pdf Download (903kB) |
Abstract
The main challenges faced by Higher Education Institutions (HEIs) in Indonesia in the context of Industrial Revolution (IR) 4.0 lie in the development of knowledge intensive skills and the development of outcome-oriented curricula. This study investigates the role of Knowledge Management Systems (KMS) and artificial intelligence (AI) in facilitating the implementation of Outcome-Based Education (OBE) as a strategic approach to improve the quality of HEIs in responding to the demands of IR 4.0. Specifically, this study investigated the knowledge creation process in KMS to support the implementation of OBE in HEIs. The objectives of this study include developing relevant knowledge pool in the area informatics for KMS, to develop a learning analytics technique for KMS in order to support the implementation of OBE in HEIs, to validate the KMS developed for OBE implementation in HEIs and to evaluate its acceptance among the users. AI in KMS is used in the knowledge creation process. The Bert2Bert model is used in the multi-document summarisation of Indonesian language knowledge in the knowledge combination process. Recommendation system on learning analysis was implemented in a hybrid algorithm combines Rule-based and Content-based filtering algorithms. KMS validation was carried out through expert assessment, and the user acceptance of the KMS was evaluated using a survey method, which adopted a questionnaire from the Technology Acceptance Model (TAM) framework. Based on the results of the knowledge creation in KMS, it succeeded in meeting learners' learning needs. The upload function represents the externalisation of knowledge. This function enables the expert to add knowledge, and followed by scraping and summarising knowledge, represented by a combination of knowledge. The scraping process extracted knowledge from online media, and knowledge from various documents or sources was then summarized. Based on the results of the evaluation of Bert2Bert model the readability for summarising 2 and 3 knowledge documents using the Flesch-Kincaid Grade Level (FKGL) showed that the average values were 20.35 and 18.1, the Gunning Fog Index (GFI) method 7.52 and 8.165, and the Dwiyanto Djoko Pranowo method 20.33 and 32.2. The evaluation explained that adults and learners at higher education levels can understand the summarized knowledge. Readability evaluation was carried out manually by Indonesian language experts. A total of 20 document knowledge was evaluated manually, and the results from document summary were understood. Internalisation of knowledge in KMS was represented through the learning analytics function, followed by an automatic recommendation system to improve knowledge. Based on the evaluation results using the confusion matrix, the Recall value was 59.1%, Precision was 100%, F1-Score was 74%, and Mean Absolute Error (MAE) of 0.97 (testing 31 data with 5 categories and target range 0-4), indicating that the recommendation system in the KMS has good classification capabilities and high accuracy in prediction. KMS also received positive validation from learning media experts, learning content, and information and communication technology (ICT) experts, with the percentage of assessment results of 79.54% and 86.1%, respectively. These results indicate the developed of KMS achieved level good and suitable for use category and does not need revision. A survey from 95 learners HEIs adopted from TAM revealed that KMS was significantly accepted for implementation of OBE via personalized learning and hence, be able to improve learners' learning. This study has significantly contributed to the development of KMS, especially in the context of personalized learning to support OBE implementation in HEIs and the integration of AI technology in the knowledge creation process in KMS.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Uncontrolled Keywords: | KMS, personalized learning, OBE, HEI, knowledge creation, summarisation, recommendation system, validation, evaluation, TAM Model |
| Subjects: | L Education > LB Theory and practice of education |
| Divisions: | School of Graduate Studies |
| Depositing User: | Muhamad Aizat Nazmi Mohd Nor Hamin |
| Date Deposited: | 20 Jan 2026 06:49 |
| Last Modified: | 20 Jan 2026 06:49 |
| URI: | http://ur.aeu.edu.my/id/eprint/1440 |
Actions (login required)
 |
View Item |
