The Arabic language and the semantic web: Challenges and opportunities

Since the advent of the Semantic Web in the late 90’s, many Web applications were created to benefit from the capabilities provided by Semantic Web technologies. These capabilities include intelligent reasoning over data, semantic search and data interoperability. However, most Semantic Web technologies are dedicated to processing Latin family scripts, thus, an apparent lack of Arabic script support in these technologies kept the research in the Semantic Web for the Arabic language untackled. This paper reports on a pilot experiment to test the support of Arabic in some of the existing Semantic Web tools, and examines the challenges and opportunities encountered when implementing Arabic Semantic Web applications. Introduction The problem with the current Web is that it is only understandable by humans. Machines cannot interpret information on the Web as people do. To illustrate this problem, suppose a person wants to search the Web for the term ‘apple’, by which (s)he means the fruit. The search engine will return results with no semantic relations. It may supply links to Web pages on ‘apple’ as in the computer company, the fruit, or even an online shop. This ambiguity in the results needs to be resolved by semantically annotating resources on the Web so that intelligent results can be retrieved. This can be done by adding an extra layer of semantics to the current Web to enable machines to understand what a Web page is about. Semantic Web (SW) technologies have proved their success in multiple domains, such as Medicine, e-Commerce, e-Learning and Biology. To extend this success to the Arabic language, a set of SW tools and applications needs to be created to fulfill the requirements of the Arabic language. Thus, the research questions we are trying to answer in this paper are as follows: • Can existing Semantic Web tools facilitate building Semantic Web applications that support the Arabic language? • What are the challenges faced by users of the Arabic language when trying to embrace Semantic Web technologies? • What opportunities will this new technology bring to users of the Arabic language? To share our understanding of the answers to these questions and to shed some light on the potential research in this field, this paper is organized as follows: in section 2, we present some background information about the SW and discuss its basic concepts. In section 3, we outline the main components of a typical SW Annotation tool. In section 4, we present an overview of SW research in Semitic languages. In section 5, we construct a simple SW application to evaluate the support of the Arabic language in some of the open-source SW tools. Finally, we conclude the paper with some of the main reasons why research conducted for Arabic has not contributed to SW research, and we provide a discussion of the potential challenges and opportunities that are awaiting researchers in this domain. Background The Semantic Web intends to improve the existing Web with a layer of machine-interpretable metadata (i.e., data about data) so that a computer program can understand what a Web page is about, and therefore draw conclusions about the Web page. The SW as defined by its creator Tim BernersLee (Berners-Lee et al., 2001), implies “... an extension of the current web in which information is given a well-defined meaning, better enabling computers and people to work in cooperation.” To add a layer of semantics to the existing Web, three challenges need to be achieved (Harmelen, 2004): The 1st International Sysmposium on Computers and Arabic Language & Exhibition 2007 © KACST & SCS The best results are given by the Okapi measure and for = 0.2. In this work we supposed that all terms of dictionaries have the same relevance . It would be interesting to calculate a specific relevance for every term in the dictionary. Then, these dictionaries (with specific relevance for each term) have to be used in the indexing stage instead of the search stage.