الفهرس 
AbstractOnly 14 pages are availabe for public view
 |  | from | 70 |  |  |
| | | from | 70 | | |
Abstract
Semantic web technologies are valuable add‐ons for digital libraries. Semantic web technologies are often deployed in digital libraries to reduce the expensive manual overhead for indexing documents, maintaining metadata, or caching for future search. However, using such techniques may cause a decrease in a collection‟s quality due to their statistical nature.
The development of digital libraries is facing some problems, like creating a standard model of digital library to understand its structure and simplify the process of creating a new digital library. Earlier attempts of digital libraries‟ models did not satisfy the need of getting standard model for digital libraries. Other problem the digital libraries face is the quality of the information retrieved by searching methods.
This thesis proposes a new model for the digital libraries, this model in an extension of the 5S formal model and the semantics is the 6th layer in the 6S proposed model of digital library. The 6S model will be Streams, Structures, Spaces, Scenarios, Societies and Semantics. In this thesis we discuss applying the semantic web technologies in digital library, the problems it faces and the importance of using ontology in digital library. Also it is displaying different models of digital library like 5S formal model. Digital libraries depend on the semantic web technologies to gain meaningful data while searching, browsing and indexing
Since the 6th layer in the proposed model depends on the semantic web technologies, and the ontology is considered an important technology in the semantic web, it is important to keep it updated with all the changes in the digital library.
A hierarchical algorithm is adapted to keep the ontology up to date with all the changes in the digital library concepts. This algorithm depends on documents keywords and naïve bayes classifier to classify books according to ontology concepts whether by creating new class or assigning it to any existing class in this ontology. Naïve bayes classifier formula is used to determine the probability distribution of a data point over all classes. It is capable of comparing not only single words, as in most current approaches, but also substrings of an arbitrary length. In this case study, we have used the digital library system that uses semantic web technology to evaluate searching using keyword based search and ontology based search as well. This system enable users to add, remove, edit documents stored in the digital library, create ontology and modify it, and also it has the ability to modify and remove metadata for documents. Ontology of documents’ subjects was created in the digital library system and is updated by the hierarchical algorithm.
The case study shows enhancements in searching results with a high precision and recall after updating the ontology with the hierarchical algorithm to include new classes classifying more documents into appropriate classes to be retrieved easily through searching with this ontology.