Abstract. A common feeling is that Semantic Web and Web 2.0 are two competing visions for the future evolution of the Web. In fact, the core technologies and concerns of these two terms are complementary. Web 2.0 applications focus on community and user interaction, which draws on Semantic Web infrastructure to facilitate information sharing. And Semantic Web provides machine-readable data to support the semantic requirement of Web 2.0. There is a significant benefit from combining Web 2.0 and Semantic Web. This paper evaluates the current development of Semantic Web and Web 2.0, and presents some approaches for compounding these two technologies.
1. Introduction
Web 2.0 is mostly a social revolution in the use of Web technologies, a paradigm shift from the Web as a publishing medium to a medium of interaction and participation (Lassila, 2007). The core technologies for Web 2.0 application involve RSS (Really Simple Syndication), AJAX (Asynchronous JavaScript and XML), microformats, and REST (Representational State Transfer). One major goal of Web 2.0 is to promote collaboration among users and improve the user interaction. Many Internet service suppliers provide applications based Web 2.0 technologies such as MySpace, YouTube, Flikr and Wikipedia to enable users to post personal content, music, video and photos on the web. Web 2.0’s success is mainly attributed to the widespread web applications such as weblog, syndication and tagging system (Mikroyannidia, 2007).
The Semantic Web with data oriented is considered as an extension of the existing Web with document oriented. The perspective of Semantic Web is to create one absolutely semantic environment which enables the agent to infer resources. The current rapid rate of growth in the amount of web information has brought a challenge to managing this information (Mikroyannidia, 2007). Most of published contents are not structured for logical reasoning, so a keyword search is not efficient any more. The Semantic Web aims to solve these problems by presenting information in machine-understandable formats that enables agents to integrate information and enhances search precision. It uses ontology to represent knowledge, however, the difficulty of ontology creation and maintenance which needs experts viewing their domains results in few Web 2.0 applications deploy the semantic data.
2. Current Work
With the past years’ development in Semantic Web since the concept came out, the Semantic Web has matured into a set of standards. Resource Description Framework (RDF) and Web Ontology Language (OWL) have become standards and new technologies are becoming maturity for embedding semantics in existing Web pages and querying RDF knowledge stores, so something exciting is clearly happening in this area (Lassila, 2007). The academic conferences have presented some current work that has been presented. Corporate research centers and academic labs are improving the vision of Semantic Web languages. W3C released the new vision of query language SPARQL in January 2008. GRDDL (Gleaning Resource Description from Dialects of Languages) and RDFa provide a technology base for making Semantic Web applications interoperate more smoothly with traditional Web applications (Hendler, 2008). Many open source software have facilitated Semantic Web development such as Jena, jBoss, PostgreSQL database and Protégé. Despite this, the current Semantic Web has no practical, scalable way for non experts to create adequate mapping functions between a large amount of constantly changing ontology and instance information or to manage their periodic maintenance (Greaves, 2007). One of six challenges (Benjamins, 2006) for Semantic Web is ontology availability, development and evolution. At the non expert level, RDF and OWL are difficult to learn and written in an accurate syntax. Most existing editors and software development environments are feasible for knowledge representation experts. The ontology creator need be familiar with complex knowledge logic. Current software developers would rather employ less semantically demanding representations, such as XML or microformats rather than RDF and OWL.
In fact, many key features of Web 2.0 applications derive from the vision of the Semantic Web. In particular, at the time the “Web 2.0” was coined the initial RDF and OWL recommendations were cleared (Greaves, 2007). Depending on a lightweight, flexible, and shared semantic model, Web 2.0 has quickly matured. Moreover, today it encompasses most exciting Web based applications: mashups, blogs, wikis, tagging systems, user-created publication systems, and social networking applications. In contrast, there are no innovative and commercially successful applications for Semantic Web. Web 2.0 has become a technology wave for software developers. The International Semantic Web Conference in 2006 began to consider encouraging Semantic Web technical capabilities to support Web 2.0 application’s semantic requirements (Greaves, 2007). However, the progress is extremely difficult. Although the basic combination reasoning is able to extend and regularize tag systems and increase search precision, the completely match between Semantic Web technologies and Web 2.0 application requirements is considerably more muddled (Greaves, 2007). Some research progresses have been implemented. A Semantic Tagging and Searching System STSS (Hope, 2007) is created to converge Web 2.0 and Semantic Web. Some approaches are used to create Semantic Web Blog.
3. Semantic Tagging
A collaborative form of this process for shared web-based resources, called “tagging” (Marlow, 2006), which is widely applied in most Web 2.0 applications. The conventional keyword based tags fail to search contents identified by different tags with the same meaning. In particular, the synonyms are not considered as the same semantic. STSS improves search results by adding context data to tags. Before posting the content to the web, users choose the defined context for the tag provided by the ontology. For example, the tag “Tiger” might have more than one context such as “animal”, “cartoon” and “story”. The user can select one or more contexts. After submitting, STSS creates a new relationship between the tag and the context selected. Adding contexts to the tags provides broader search results, because the search can return all content that matches the keyword and all content associated with tags that are context of the keyword. The benefit of this approach allows the user easily to edit tags without destroying the ontology consistencies (Hope, 2007). However, due to the only a single ontology used in the system the contexts cannot be shared with other ontologies.
The other approach to solve the problem of tagging is to make explicit semantics behind the tag space in social tagging systems (Specia, 2007). Compared with the approach above, it does not rely on the additional context. This solution consists of three steps: pre-processing, clustering and identifying concept and relationship. The first step involves removing unusual tags which cannot be found in ontologies, and grouping similar tags. Each group is replaced by its representative. The second step is to analyze the tag space to indentify groups and finally generate related cluster (Specia, 2007). In the third step, the knowledge provided by other sources including ontologies available on the web is implemented to identify relationships between tags in each cluster, if they exist categorizes them. After that, the tags will be mapped into concepts, properties and instance of ontologies. This approach is feasible and promising, which makes use of the existing ontologis searched by Swoogle. The new ontology derived from folksonomies (Specia, 2007) provides machine-readable data, which enables Web 2.0 search engines to retrieve more accurate information.
4. Semantic Weblog
(Karger, 2004) constructs a semantic blogging environment based on a Semantic Web browser “Haystack”. The most current blogs have machine readable contents encoded in an XML format called RSS which allows individual blogs to be aggregated together. In this paradigm, the XSL transform is considered as a standard to translate existing RSS 9.0/2.0 feeds into RDF. In this model, Haystack provides specialized forms retrieving different level of semantics for RDF representation automatically and enables users to publish such information using RSS 1.0 (Karger, 2004). Another scenario of Semantic Weblog is presented in (Ankolekar, 2007). A plug-in that uses Semantic Web technologies allow bloggers to add information about blog entries. The semantic data can be created by a light weight ontology language FOAF which is easier to handle than complex RDF/OWL.
5. Future Work and Conclusion
Form the view of the business model, in order to prosper in the market the Semantic Web should provide an essential capability better, faster, or cheaper for no expert users than any other competing technology. Web 2.0 applications have made it easier for everybody to share a wide variety of contents over the Internet. It is necessary to deploy an efficient and accurate searching method for specific content in this ocean of information. Semantic data creations tools will help embed the semantics to web information. Web 2.0 and Semantic Web need complement each other, and in fact both communities need elements from each other’s technologies to overcome their own limitations. One believable thing is that semantic technologies will bring a robust and extensible enhancement for emerging Web 2.0 applications. Definitely, data reuse, distribution and aggregation can be greatly facilitated by the Semantic Web’s infrastructures. Web 2.0 proponents should encourage themselves to embrace those Semantic Web technologies. In the future, the two communities are expected to create an innovative Web application as the representation of Web 2.0 and Semantic Web by exploiting each other’s achievements and insights.
References
Ankolekar,A. Krotzsch,M. & Tran,T. 2008, ‘The tow cultures: Mashing up Web 2.0 and the Semantic Web’, Elsevier Science Publishers B. V, vol.6, no.1, pp.70-75.
Benjamins,R,V. & Contreras,J. 2006, ‘Six Challenges for the Semantic Web’, [Online] Available at: http://www.cs.man.ac.uk/~ocorcho/documents/KRR2002WS_BenjaminsEtAl.pdf
Greaves,M. 2007, ‘Semantic Web 2.0’, IEEE Educational Activities Department, vol.22, no.2, pp.94-96.
Hendler,J. & Golbeck,J. 2008,’Metcale’s Law, Web 2.0 and Semantic Web’, Elsevier Science Publishers B.V, vol.6, no.1, pp14-20.
Hope,G. Wang,T. & Barkataki,S. 2007, ’Convergence of Web 2.0 and Semantic Web: A Semantic Tagging and Searching System for Creating and Searching Blogs’, Proceedings of the International Conference on Semantic Computing, pp.201-208
Karger,D.R. & Quan,D. 2004, ‘What Would It Mean to Blog on the Semantic Web?’, Proceedings of the Third International Semantic Web Conference (ISWC2004), pp.214-228.
Lassila,O. & Hendler,J. 2007, ‘Embracing Web 3.0’, IEEE Educational Activities Department, vol.11, no.3, pp.90-93.
Mikroyannidia,A. 2007, ‘Toward a Social Semantic Web’, IEEE Computer Society Press, vol.40, no.11, pp.113-115.
Marlow,C. Naaman,M. Boyd,D. & Davis,M. 2006, ‘HT06, tagging paper, taxonomy, Flickr, academic article, to read’, Proceedings of the Seventeenth Conference on Hypertext and Hypermedia, pp.31-40.
Specia,L. & Motta,E. 2007, ‘Integrating Folksonomies with the Semantic Web’, [Online] Available at: http://www.eswc2007.org/pdf/eswc07-specia.pdf