University of Macedonia Economic and Social Sciences
Αυτόματη παραγωγή και διαχείριση μεταδεδομένων πολυμεσικού περιεχομένου για την υποστήριξη συστημάτων ηλεκτρονικής εκπαίδευσης
Abstract
dc:descriptionThe search for information in the World Wide Web and the discovery of the desirable information resources in our modern days where people are overwhelmed by innumerable quantities of information which most of the times do not coincide with their information needs can only be achieved if resources have been adequately described in order to correspond to complex and diverse search criteria. Especially in an educational environment where the amount of available educational resources is growing rapidly while their kinds and formats have acquired varied dimensions (including all sorts of multimedia information) finding suitable educational resources is no more an easy task. Metadata is the key to find desirable resources. It is a term of the Internet era for the information that librarians traditionally have been using in their book catalogues and usually refers to descriptive information for information resources contained in physical or electronic means. Metadata constitute the cornerstone of the Semantic Web i.e. the expansion of the current Web in which information acquires a clearly defined meaning which allows computers and humans to cooperate harmonically. Focusing more specifically on educational resources the demand for their adequate description can be satisfied by utilizing extended data description systems (educational metadata schemas). However the process of creating metadata is problematic and time consuming owing to the complexity of the schemas and the required knowledge background but also to the human effort required for creating complete and correct metadata. This dissertation has the goal to study ways and methods to solve this category of problems related to the creation of educational metadata aiming at contributing to the research field of automatic metadata generation for educational resources. In order for this goal to be achieved a research plan was drawn including a thorough study and review of the bibliography covering the research field under investigation. Hence the concept of learning object which constitutes the new trend in the format in which educational resources are being developed is studied and analyzed. Moreover the metadata standard for learning objects IEEE LOM is examined thoroughly. The problem of automatic metadata generation is put under investigation and sources for metadata creation are distinguished in the content context use and related content of the resources. As a result of the conducted study an innovative methodological approach for the automatic generation of metadata which exploits the relations among the resources to be described is introduced. After a slight adjustment of the semantics of LOM relations the proposed methodology is applied to this metadata schema producing 72 logic rules for the automatic generation of metadata of a learning object from the already existing metadata of its related objects. A practical implementation of the proposed approach is performed by the development of an integrated LOM metadata management system which in addition to its editing capabilities of LOM metadata it incorporates the mechanism for the automatic generation of metadata by applying the logic rules created before. The operation of the system proves that it can be a powerful and credible tool for the automatic generation and management of metadata of educational resources which can promote the goal of easy searching and finding suitable resources.
Degree
thesis:*- Grantor dc:publisher
- University of Macedonia Economic and Social Sciences
- Year dc:date
- 2007
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Margaritopoulos, Merkourios
- Μαργαριτόπουλος, Μερκούριος
Subjects
dc:subject × 14- Μεταδεδομένα
- Αυτόματη παραγωγή μεταδεδομένων
- Μεταδεδομένα μαθησιακών αντικειμένων
- Επεξεργαστής μεταδεδομένων IEEE LOM
- Λογικοί κανόνες
- Metadata
- Automatic metadata generation
- Learning object metadata
- IEEE LOM metadata editor
- Logic rules
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/24758