Semantic Web

The Semantic Web is the future of the World Wide Web. The Web that we know and use today is designed for human utilization only, but not for the computer process itself. The Semantic Web will add more functionality to the Web pages in terms of making it easy for software agents to do their tasks (Berners-Lee, 2001).

This means that semantics make it possible for the Web page to understand and comply with the needs of people or machines on its own. A Semantic Web is not a separate Web page but something that needs to be encoded into the system for the easy use of the people and the computer.

Recent developments in Semantic Web states that “it will usher in significant new functionality as machines become much better able to process and understand the data that they merely display at present” (Berners-Lee, 2001).

The vision of Semantic Web is to present data on the Web for application and not simply just for display purposes. Using it will enable computers and other machines to understand and respond to semantic documents and data.

There are current tools for creating and developing Semantic Web. The list of tools just keep growing and growing everyday. The most common are those such as RDF, RDFS, and OWL.

There are different tools for the Semantic Web such as tools to ask described information, tools to have detailed classifications and characterization, tools to define logical relationships, and tools to extract and bind traditional data sources.

Currently, there are two very important technologies behind the development of the Semantic Web. These are the XML or Extensible Markup Language and the RDF or Resource Description Framework.

The use of XML gives users the chance to create tags or labels on Web pages. It allows Web users to “add arbitrary structure to their documents but say nothing about what the structures mean” (Berners-Lee, 2001).

RDF means Resource Description Framework. It is one of the fundamental building blocks of the Semantic Web. It is responsible for the exchange of data on the Web. It also facilitates the merging of data and supports the schemas. RDF's “triple” model, which links the structure of the Web to the URIs or Uniform Resource Identifiers, allows data to be mixed and shared in different Web applications (Herman, 2008).

In addition, according to Herman (2008), RDF is a “language for representing information about resources in the World Wide Web.” It is usually used for information that needs application instead of being displayed in a Web page. Because it is considered a language, an RDF triple is compared to a subject, verb, and object in a sentence that can be written with XML tags.

Just like an ordinary sentence, the subject, verb, and object or the triples of the RDF are recognized by a URI. A common example of a URI is the URL or Uniform Resource Locator. These triples create data in the Web. The URI makes sure that the data are accessible to anyone who browses the Web (Berners-Lee, 2001).

RDFS or RDF Schema is the vocabulary description language of RDF. An ordinary RDF cannot provide description for properties or its relationship to other resources. But an RDFS can define and describe classes, properties, and other resources. It is a “semantic extension of RDF written in the RDF way” (Herman, 2008).

Herman (2008) added that “the RDF vocabulary description language class and property system” is the same as programming languages such as Java. The only difference is that the RDFS “describes properties regarding the classes of resource to which they apply.” Resources of the Web are divided into classes. Members under a class are termed instances. It is stated that “a class may be a part of its own class extension and maybe an instance of itself.” Aside from classes, there are also properties wherein the concept of an RDF property is described as a link between the subject and object in the RDF.

Schema Properties

Another application designed to develop Semantic Web is the OWL or Web Ontology Language. Ontology is the representation of terms and their interrelationships. The OWL is better than XML, RDF, and RDFS in terms of presenting more vocabulary and formal semantics. It has three sublanguages namely OWL Lite, OWL DL, and OWL Full. This application is best used when the information needs processing and not just as display purposes (Herman, 2008).

This Web Ontology Language caters to the needs of the users and developers depending on their purpose and support system. People who intend to use OWL are given the opportunity to consider and choose which type of OWL satisfies their wants and needs.

The three sublanguages of OWL also have their specific purposes. The OWL Lite supports users who need “classification hierarchy and simple constraints”. OWL DL supports users wanting the “maximum expressiveness while retaining computational completeness and decidability.” OWL Full supports users who are in need of “maximum expressiveness and the syntactic freedom of RDF with no computational guarantees (Herman, 2008).

One of the current available applications for Semantic Web tools is Protégé. It is a free, open-source ontology editor written in Java. It is a knowledge-based framework developed by Stanford University. Protégé has a plug-and-play function which characterizes it as an extensible and flexible development of ontologies. It is mostly used by developers and users for knowledge solutions in fields such as biomedicine, intelligence gathering, and corporate modeling (Burleson, 2007).

There is an extension of the Protégé which is the Protégé-OWL editor. It supports OWL, the tool considered as the most recent and better than the others. Protégé-OWL makes it possible for users to load and save OWL and RDF ontologies, edit and visualize classes and properties, define logical class characteristics as OWL expressions, execute reasoners, and edit OWL individuals for Semantic Web markup (Herman, 2008).

Semantic Frameworks

The Semantic Web Framework is the “structure where Semantic Web applications are organized and developed.” There are principles as to how the framework should be. First, the framework must be developer-oriented. It must consider all users whether with Semantic Web knowledge or not. Second, the framework must be easy to understand. The components must be organized according to class and property. Third, the framework must be inexpensive to adopt. Application or system upgrades of the Semantic Web should be easy. Fourth, the framework must be semantics-focused. Because the users are dealing with developing a Semantic Web, it is best to focus on it. Fifth, the framework must be component-based for better implementation and functionality of the Semantic Web. Sixth, the framework should be evolving. It means that it should allow new components to enter and be open to modifications to ensure that the Semantic Web is developing in time (Garcia-Castro, 2008).

The purposes of a Semantic Web framework includes describing the types, functions, and dependencies of Semantic Web technologies, facilitating technology reuse, presenting how to obtain interoperability with Semantic Web technology, and helping make decisions in developing Semantic Web applications (Garcia-Castro, 2008).

There are many frameworks available for the development of the Semantic Web. An example of a Semantic Web Framework is Jena. It is a Java framework that creates and develops Semantic Web applications. It can do programming for RDF, RDFS, and OWL. It is open-source and includes a rule-based inference engine. Its advantages include an RDF API, an OWL API, able to read and write RDF in RDF/XML, N3, and N-Triples, and in-memory storage (jena.sourceforge.net).

Another example is CubicWeb. It is a Semantic Web application framework that provides users to develop web applications by reusing components known as cubes. Its advantages include an engine driven by the data model, a query language, a selection+view mechanism, a wide variety of reusable components to choose from, the capabilities of the Python programming language, and the reliability of databases, directories, and storage (www.cubicweb.org).

Below is the image showing the different components of a Semantic Web Framework. There are seven dimensions and piled above them are the corresponding components.

Components of the Semantic Web Framework

The intention of the Semantic Web Framework is to help developers in building inexpensive Semantic Web applications. It is a useful reference for further development of the technology. Uses of this framework include the proper knowledge of the components needed in application or those to be reused (Garcia-Castro, 2008).

According to Garcia-Castro's research team (2008), there is an extended use of the Semantic Web Framework adds a new dimension for placement of social components. There will be development of specifications for the components, their life cycle, and interfaces. New guidelines will also be implemented for these components and for reusing them during Semantic Web application developments.

Framework Data Flowchart

An ontological Semantic Web Framework gives users access to the knowledge base. They have the chance to assess and evaluate the contents, see if the resources are important for the development, and to mix and exchange data. Ontologies are the providers of vocabulary for the integration of knowledge bases, data format, and database schemas used in the Semantic Web (Yablonsky, 2009).

Semantic Web Trends, Challenges, Future, and Requirements

As mentioned earlier, the ordinary Web that we know and use today lacks something in terms of technological advancement. The Semantic Web is the solution to that lack. It has been developed to fill what the Web cannot do. That is to make contents, information, data, and other documents available for automatic processing by the computer itself.

The Semantic Web has many benefits such as consistent interconnection and minimal error messages, and will be distributed to everyone just like the ordinary Web. It can be accessed by corporations and individuals (Berners-Lee, 2001).

Two years ago, Cardoso (2007) made a study on the trend of Semantic Web. Of the 627 participants that were asked where they represent knowledge with ontologies, 31 percent answered education. 25.5 percent was for computer software, 17 percent for the government, another 17 percent for business, and 16.5 percent for life sciences. Only 13 percent answered for communication, 12.8 percent for media, and 11.3 percent for healthcare.

Web is the use of ontologies

Another trend in Semantic Web is the use of ontologies. Out of the more than 600 participants, almost all revealed that they use ontologies for sharing of information among other users and software agents. It means that data can be understood by humans and the computers. Only 12 percent use ontologies for other purposes such as code generation, data integration, data exchange, document annotation, information retrieval and many more (Cardoso, 2007).

Use of available languages for developing Semantic Web ontologies was also discussed. More than 75 percent of the participants use the most popular and better tool which is the OWL. According to the participants, the smallest ontologies that they created had less than 100 concepts and the largest had more than a hundred to a thousand concepts (Cardoso, 2007).

The primary goal of the Semantic Web is to develop languages to describe information that is understandable by machines. If there is an opportunity for the Semantic Web to be fully used worldwide, there are challenges that await the technology. Some of the current challenges are the development of ontologies, formal semantics of Web language, and the trust and proof models (Lu, 2002).

Ontologies are the most essential part of developing the Semantic Web. There are different aspects of this part which include ontology representation languages, ontology development, ontology learning approaches, and ontology library systems. These aspects manage, adapt, and standardize the ontologies (Lu, 2002).

Managing ontologies is difficult since it is the core of Semantic Web. There must be an organized library system with open storage, classification, and versioning. Ontologies must be identified and organized carefully since it changes over time. Because ontologies evolve, constant upgrades and extensions must be used. Proper adaptation of the ontology is important in developing the Semantic Web. After managing and adapting to the ontologies, making it standard is the next step. Choice of ontology representation language is needed depending on the needs of the user (Lu, 2002).

Structure of the Semantic Web

The structure of the Semantic Web has three layers namely the metadata, schema, and logical layers. The tool considered for the metadata layer is the RDF and RDFS for the schema layer. But the RDFS lacks formality and one of the challenges of the Semantic Web is to reconstruct it to a formal semantics. Formal semantics can lessen the confusion and other problems attributed to the development of Semantic Web using languages and tools (Lu, 2002).

As soon as the Semantic Web is open for the use of the public worldwide, it would expose everything to anyone. The challenge of security comes in. There should be a use of proofs such as digital signatures to check the sources. Security systems like encryption and access control should be able to protect the users as well as the machine. Users must have a hierarchy of trusted network sources on the Semantic Web (Lu, 2002).

In order for the Semantic Web to be successful, the computer must have access to the information they will use. This is known as knowledge representation technology. Continuous upgrades lead to slower effectiveness of the Web service. The new prerequisite in developing Semantic Web is to be able to provide layers that can let people use the technology (Biskup, 2004).

The new requirements of a better Semantic Web include core support for RDF, RDFS, and OWL, support for SPARQL, be able to store large RDF datasets, be able to provide inference capabilities for OWL, be able to organize linked data, be able to apply role-based security, be able to publish SQL databases and LDAP repositories as RDF/OWL linked data, and to be able to extract semantic metadata from sources (Biskup, 2004).

In terms of the future, Cardoso (2007) believed that the Semantic Web will penetrate the mainstream in five to 10 years from now. But the participants who are Semantic Web users and developers believe that since Semantic Web is beneficial, it could be out for worldwide public use in less than two years.

In a simple definition, a Semantic Web is a Web that is enhanced to be able to clearly understand the contents and structure of a web page. It makes web search engines and web browsers capable of reading and processing web contents to provide better services to users. It will enable computers and other machines to process ideas and solve difficult optimization problems.

As of now, the tools for developing Semantic Web are available such as RDF, RDFS, and OWL. It would still be after a few years for the Semantic Web to fully conquer the world.

Arabic Language Challenge

The Semantic Web will soon conquer the world of technology. But the tools and languages are not ready for Arabic language users. The rise and development of the Semantic Web must address the challenges of the complicated Arabic language.

In the discussion by Al-Khalifa and Al-Wabil (2007), they questioned the existing Semantic Web tools and applications in supporting the Arabic language. Since the technology would become available worldwide, it must prepare itself for the hundreds of different languages to support.

There are issues concerning the Semantic Web that pose challenges to the Arabic language and its users. There is lack of Arabic language support in the available and current Semantic Web tools, lack of existing Arabic Semantic Web applications, and limited support for Arabic research on Semantic Web technologies (Al-Khalifa, 2007).

A common problem with Semantic Web tools is the processing and encoding of Arabic texts. Currently, there are a lot of applications that can encode Arabic scripts. To solve the problem, Semantic Web tools should focus on one encoding schema for Arabic such as Unicode (Al-Khalifa, 2007).

Another problem with the Semantic web is that 49 percent of the ontologies in the ontology libraries are written in English. This is in connection with the lack Web tools and software development applications that cater to the Arabic language and its users. If there are tools and applications that support Arabic, it is only very limited and still lacking the full functionality and processing (Al-Khalifa, 2007).

Distribution of Languages for Ontology Creation

In terms of research in this technology, there is limited support for Arabic academics. The outcome of academic researches in the field of Semantic Web technologies are the tools and applications as a result of the study. Aforementioned applications such as Protégé, Jena, and CubicWeb are just some of the end products of researches and studies (Al-Khalifa, 2007).

Conducting Arabic researches and studies on this topic are very limited due to the lack of resources, skills, funding, and interest or enthusiasm in this field. The Arab community must be interested as well as eager to allocate funds and resources in order to solve this problem (Al-Khalifa, 2007).

In order to gain interest, the community must be exposed and should have experiences with different Web technologies. They need funding for the cost of Semantic Web application development and maintenance.

There are also other challenges that Arabic language users could encounter if they come across tools and applications that cater to their language. There can be a need for further development of Arabic vocabularies and ontologies to be able to process semantics better. This vocabularies and ontologies would help the Arabic language users to communicate with the machine properly and support the exchange of data. An example of this development would be the creation of the Arabic WordNet (Al-Khalifa, 2007).

Another challenge is the difference in the computational processing of Arabic and English. Since Arabic is a complex language in terms of grammar, morphology, and semantics, tools used for English cannot be easily and properly used for Arabic. Because the Semantic Web has the process of information extraction and data exchange from documents and other content, tools are needed to be able to process the Arabic language (Al-Khalifa, 2007).

Aside from the challenges mentioned, there are a few more factors that affect the usage of Arabic. It includes the fact that Arabic is very filled with variety and derivations. It makes morphological analysis of the language difficult task. Also, in Arabic vowels, the lack of accent marks in the written texts makes it vague and the difficult morphological analysis is needed (Al-Daimi & Abdel-Amir, 1994).

The most common problem with Arabic is the writing direction. It goes from right to left and some of the characters in the Arabic alphabet get distorted depending on its location in the word. Capitalization is one of the problems as well. Arabic does not use capitalization in its language and it makes it difficult for readers to determine which are proper names, abbreviations, or acronyms (Al-Daimi & Abdel-Amir, 1994).

Screenshot of Arabic Protégé Application

Conclusion

This paper discussed, in an organized manner, what the Semantic Web is all about. It talked about the different tools and application used for this fast emerging technology. The frameworks and challenges are also presented to prepare the readers for what is to come. This paper presented how the Semantic Web is affecting or going to affect the world in the future especially in terms of language usage specifically Arabic.

A great and clear example of what the Semantic Web can offer is the “plug-and-play” technology. In the past years, installation and configuration of software or other applications are manually done in the computer. The “plug-and-play” technology lets the user to just plug something in the computer and it will automatically be read. Same thing goes with the Semantic Web. The computer itself will be the one to process the things needed by the user.

The Semantic Web is an example of how fast technology can change. It is not just a tool for people to use but it can assist in evolution of knowledge as well. The Semantic Web, when it opens itself for public use worldwide, would be exposing new concepts and would let everyone exchange expressions. Its use of a unified logical language will enable the computer to connect the world to a universal Web.

By this connections and links, humans can have access to a wide array of knowledge and ideas. In this way, all people can live together, work together, and learn together. It opens a lot of possibilities for the next generation of technology users.

That is why it is very important for the Semantic Web tools and applications to prepare themselves to support all languages such as Arabic in order to fulfill the Semantic Web goal of connecting the world into one network.

References

Al-Daimi, K. & Abdel-Amir, M. “The Syntactic Analysis of Arabic by Machine.” Computers and Humanities 1994. Vol.28, No.1.

Al-Khalifa, Hend S., and Areej S. Al-Wabil. "The Arabic Language and the Semantic Web: Challenges and Opportunities." The 1st International Symposium on Computers and Arabic Language & Exhibition. London. 2007. Lecture.

Berners-Lee, Tim, James Hendler, and Ora Lassila. "The Semantic Web." Scientific American May 2001. Print.

Biskup, T., and J. M. Gomez. "Building Blocks of a Semantic Web Framework: Requirements Analysis and Architectural Implications." Database and Expert Systems Application (2004). Print.

Burleson, Cody. "Introduction to Semantic Web Vision and Technologies." Semantic Focus. 29 Oct. 2007. Web. <http://www.semanticfocus.com/blog/entry/title/introduction-to-semantic-web-vision-and-technologies-part-4-protege-101-screencast/>.

Cardoso, Jorge. The Semantic Web Vision: Where are We? Diss. University of Madeira, 2007. Print.

"CubicWeb - The Semantic Web is a Construction Game." CubicWeb. Web. <http://www.cubicweb.org/>.

Garcia-Castro, Raul, Asuncion Gomez-Perez, and Oscar Munoz-Garcia. The Semantic Web Framework: A Component-Based Framework for the Development of Semantic Web Applications. Diss. Universidad de Politecnica de Madrid, 2008. Print.

Herman, Ivan. "W3C Semantic Web FAQ." W3C Semantic Web. 27 June 2008. Web. <http://www.w3.org/RDF/FAQ>.

"Jena - A Semantic Web Framework for Java." Jena. Web. <http://jena.sourceforge.net/>.

Lu, Shiyong, Ming Dong, and Farshad Fotouhi. "The Semantic Web: Opportunities and Challenges for Next-Generation Web Applications." Information Research 7.4 (2002). Print.

Yablonsky, Sergey. Semantic Web Framework for Development of Very Large Ontologies. Thesis. St. Petersburg State University, 2008. 2009. Print.