| Abstract |
A bibliographic base contains information pertaining to a collection of documents (title, authors, date of publication, publisher, etc.) and is used for their organization and retrieval. Such a base could contain annotations relating to the documents' contents. The value of keeping such annotations in a database lies in that they can be shared by the members of a group, and can form the basis for document retrieval and evaluation. This thesis deals with the representation and management of a bibliographic base and the annotations contained therein. The aim of the project is the development of a document annotation system that supports creation and management of annotations on documents. The system described is intended to be used by either a single user or a group of co$-$workers, in order to assist in the organization, sharing and retrieval of documents and their associated annotations. Information is described using the knowledge representation language Telos. A representation model for the logical structure of articles and books was implemented in accordance with the ISO 12083 standard. Mechanisms for navigating within the logical document structure were developed, as well as methods for interconnecting the logical structure with the content which is stored externally with respect to the database. Following this a representation model for the annotations which can accompany a document or a part thereof was implemented. Specifically, following a literature study and discussions with researchers, we define the points of view with which annotations are made (e.g. subject classification, critical assessment, relationships with other documents, relationships with projects, etc.), as well as the classes corresponding to the annotations' content. The operational kernel of the Document Annotation System is the Semantic Index System (SIS), developed at the Institute of Computer Science, Foundation for Research and Technology~-~Hellas. In order to enhance the efficiency of user~-~system interaction, a set of predefined queries was developed, taking into consideration the basic notions and their interrelationships. For the update operations we have customized the SIS Entry Forms tool to our data base model. Throughout the course of designing and implementing the system, feedback from individuals who used it was received, leading to the improvement of the system.
|
Σχεδίαση και Υλοποίηση Ενός Συστήματος Υπομνηματισμού Εγγράφων
