The Semantic Web- Are We There Yet

Lecture Notes in Computer Science: The Semantic Web: Are we there yet?

Roushdat Elaheebocus School Of Electronics and Computer Science University of Southampton, UK [email protected]

Abstract. In this paper, we have taken a look at the use of semantics applied to the context of hypertext before the web became widespread. Also we covered the main visions of the semantic web pioneers. We end up with an analysis about the current situation of the semantic web;To what extent the goals have been achieved and the challenges we are facing.

Keywords: semantic web, hypertext, current state, challenges

1 Introduction In the context of hypertext, the W3C [1] defines the semantic web as providing "a common framework that allows data to be shared and reused across application, enterprise, and community boundaries. And is "based on the Resource Description Framework" to enable not only humans but also machines to understand contents available from different online sources cooperating with one another [2]. It has been proposed by Tim Bernes Lee [3]as an extension of the Web which is itself, one that links documents whereas the former links data [4].

2 Early uses of semantics in pre-web hypertext Semantics have been the subject of research much before the web was invented. As from the year 1987, Frank G. Halasz et Al [5] proposed the idea of interlinking electronic cards NoteCards creating a semantic network. Developed at Xerox PARC, it has been a famous hypertext system mainly because of its extensive documentation by the researchers.

In 1993, D.H Jonassen and Sh.Wang [6] carried out experiments on mapping expert knowledge structure onto hypertext which results into some kind of structure that can then be inter-linked with other networks, therefore forming a semantic network. H. P. Frei and D. Stieger [7] have used to use the relationship of semantic links between nodes to implement retrieval algorithms capable of extracting relevant information from large collection of data stored in the form of hypertext or hypermedia. All these works have definitely contributed towards the envisioning of the extension of the Web with semantics which we know today as the Semantic Web or Web 3.0.

3

Visions of the Semantic Web

The main objective behind this Semantic Web initiative is to solve the problem of information overload that we are facing by adding intelligence to the web [8]. Some years back, a temporary remedy was found: Search Engines. However as the amount of information being made available online keeps growing, conventional search engines are unable to cope since they rely on keywords without knowledge about the meanings of the information they crawl through. Hence, to provide information consumers with the most relevant data, the Semantic Web which deals with meanings and relationships can help. Web Services that have already become widespread on the Internet these days are predicted to become even smarter allowing users to 'discover' them dynamically and select the most appropriate one given a set of parameters and also compose new services on the fly through the combination of two or more existent services [9]. The pioneers of the Semantic Web envisions it as a virtual field of information with agents working their way around, picking up what their masters require. These agents are described as being intelligent enough to make decisions, learn from their experiences and coordinate their actions among other agents [3, 10]. Researchers have even proposed an extension to the semantic web which they called the Conceptual Web [11]. Built on top of the Semantic Web, this extension will add the capability of making information available to users in an appropriate form and can be achieved through the use of “UML and a technique called conceptual browsing".

4

Where we stand today and challenges that we face

According to Jorge Cardoso in his survey of entitled "The Semantic Web Vision: Where Are We?" [12], developers of the Semantic Web have opted for two main technologies to represent knowledge namely; OWL and RDF Schema, the former being more popular than the latter. Also 70% of the survey participants planned to release real-world system within two years (2008) which means that by now we

already have numerous semantic web systems in operation although most run in the background. One example will be the mobile phone company Vodaphone that has adopted the use of semantics to enhance user experience in terms of searches on its website. A simple way of adding meaning to data has been started to be used in Web 2.0 in the form of metadata called tags. Through time, tagging has been improved and we now have collaborative tagging whereby users can add tags, considerably alleviating the burden from content providers and also enhancing metadata in terms of quantity and quality. However, this can be considered as a double-edged sword since allowing anybody to add tags introduces the risk that some people with malicious objective can add completely irrelevant and misleading tags. We can refer to this as "dirty-tagging" which in fact, effectively suppress the usefulness of tags. The Semantic Web has made it possible to get around this problem by introducing the concept of automated semantic tagging [13] and tags can thus be automatically generated from data. However the use of automated tagging has not been widespread so far. One example will be Twine [14] which uses algorithms to generate tags. E-Commerce sites have morphed from mere conversion of real-to-virtual conversions to rather sophisticated shops of their own with a lot of enhancement thanks to technologies brought forward by web 2.0 such as XML. What we have not seen though is automated match-making for buyer-seller service. Research in applying semantic web technologies to help in this context is ongoing. For example, David Trastour et al's work on building an automated matchmaking and negotiation service for e-commerce [15]. Applying David's research which seems to be centralised to Tim's concept of agents [3], we can obtain independent and robust agents. Empowering users has been a goal of both Web 2.0 and the Semantic Web. While Web 2.0 has achieved part of it through community-based aspects such as commenting, rating and collaborative applications, Semantic Web pushes the limit even further by giving users more control over the content presentation. This concept has been proposed in the form of a semantic browser [16, 17] and relies on the fact that users have direct access and control over the data in their client thus allowing them to manipulate it. Currently, technologies such as mashups, introduced in Web 2.0 can be said to be doing part of the job by enabling the integration of data from different sources into a single applications through the use of RSS feeds. One good example of this technology being used is Yahoo Pipes [18]. Social Networking sites such as Orkut, Hi5 and Facebook appeared on the web around year 2002 and have been considered as Web 2.0 products. However these sites are now embracing the semantic web by using the Resource Description Framework (RDF) to publish their members' profile [19]. Although few tools exist to make use of these semantic data, they will be in then near-future. The success of the World Wide Web has been primarily due to its simplicity and therefore has been more readily embraced by the online community [20].And for the Semantic web to be similarly accepted and become as successful as the Web, according to Brian McBride's view [21], it needs to 1. "emphasize practical applications" whereby different stake holders can clearly see the benefit of such technologies in areas such as business applications and

"Immediate gain needs to outweigh extra cost" [20] so that content providers can see the benefit. 2. "software must be simple and tolerant of error", the very aspects that were primordial in the success of the web; simplicity and fault-tolerance. 3. "develop applications now" suggests that we have enough technologies such as RDF, OWL and SPARQL to enable the development of applications that can better demonstrate the usefulness of semantic web. 4. "Develop a common open source infrastructure of high quality References" will enable cost effective adoption by the industry as well as innovative ideas to make it through from a large community of researchers and developers. According to Veltman [22], a major challenge is that different groups has their own view about what the semantic web is about and he goes further to describe four major groups, namely; World Wide Web (W3), Dublin Core, a small group within the AI community and 'cultural semantics' . So we need to settle for an agreed definition of what the Semantic Web is. Another barrier in the adoption of the Semantic Web is the "chicken-and-egg problem" described by David Huynha et al [17] as the situation whereby there is not enough semantic data upon which users can directly use tools and from content providers' view, there is not enough information consumers for semantic data, therefore no benefit for them to provide it. This issue has also been raised by V. Richard Benjamins et al [23]. As a possible but temporary solution, semantic web browsers were proposed [16, 17]. But this still has to prove its effectiveness in getting around the problem, which it has not yet. Scalability could be a problem if semantic data is not properly organised and stored. Like the original web, if we want the Semantic Web to reach the entire globe, it should be made scalable. Other challenges described by Benjamins et al [23] include the ability of the semantic web to interact with data in different languages, increased intuitiveness by users when working with data through innovative information visualisation techniques and also standardisation of the different technologies by the W3C [1]. However, at the time the paper was published, in the year 2002, the semantic web was still very young. Six years later, we can clearly see that most of the challenges have been, in not fully, at least partially been overcome. And as the technologies become more mature, we will witness more standardisation in place.

5

Conclusion

Definitely the Semantic Web has a bright future ahead. With now larger companies joining in, we have witnessed a similar trend in the adoption of its predecessor, the Web; a handful of scholars, then startups and small companies and when the big ones join in, we reach a level of successful embracement of the new technology. Some researchers [24] argue that in order for the semantic web to be accepted, a 'killer app'

must be produced from it that will act as torch bearer for the Semantic Web and this application is not here yet or maybe it is but we do not know about it just yet. The question that remains to be answered: Will we be able to overcome the challenges that remains in the semantic web or will we need a web 4.0 (assuming the semantic web is web 3.0) which will be even more intelligent?

6

References

1 “W3C Semantic Web Activity,” http://www.w3.org/2001/sw/ .Accessed on 1 November 2008 2 Hendler, J., Berners-Lee, T., and Miller, E. Integrating Applications on the Semantic Web, 2002, http://www.w3.org/2002/07/swint. 3 Tim Bernes-Lee, James Hendler, and Ora Lassila, “The Semantic Web: Scientific American,” Scientific American Magazine, May 17, 2001, http://www.sciam.com/article.cfm?id=the-semantic-web&print=true. 4 “W3C Semantic Web FAQ,” http://www.w3.org/2001/sw/SW-FAQ.Accessed on 29 November 2008 5 Frank G. Halasz, Thomas P. Moran, and Randall H. Trigg, “Notecards in a nutshell,” in Proceedings of the SIGCHI/GI conference on Human factors in computing systems and graphics interface (Toronto, Ontario, Canada: ACM, 1987), 45-52, doi:10.1145/29933.30859, http://portal.acm.org/citation.cfm?doid=29933.30859. 6 D. H. Jonassen and Sh. Wang, “Acquiring Structural Knowledge from Semantically Structured Hypertext.”, http://tecfa.unige.ch/staf/staf9698/mullerc/3/jonassenWang.html .Accessed on 1 November 2008 7 H. P. Frei and D. Stieger, “Making use of hypertext links when retrieving information (preweb semantics),” in Proceedings of the ACM conference on Hypertext (Milan, Italy: ACM, 1992), 102-111, doi:10.1145/168466.168502, http://portal.acm.org/citation.cfm? id=168502&dl=. 8 Oren Etzioni, Mike Perkowitz: “Adaptive Web Sites: an AI Challenge”, IJCAI-97. http://dli.iiit.ac.in/ijcai/IJCAI-97-VOL1/PDF/003.pdf 9 Ankolekar, A., et al., (2001), "DAML-S: semantic markup for web services.", in: The First Semantic Web Working Symposium, pp. 411-430, Heidelberg: Springer-Verlag Heidelberg. 10 Payne, T., Singh, R. & Sycara, K. (2002), "Browsing schedules - an agent-based approach to navigating the semantic web", in: The First International Semantic Web Conference (ISWC), Sardinia, Italy. pp. 469-474 Heidelberg: Sprnger-Verlap Heidelberg. 11 Ambjörn Naeve et al, “The Conceptual Web – our research vision,” http://cid.nada.kth.se/pdf/CID-156.pdf. 12 J. Cardoso, “The Semantic Web Vision: Where Are We?,” Intelligent Systems, IEEE 22, no. 5 (2007): 84-88, doi:10.1109/MIS.2007.4338499. 13 Stephen Dill et al., “SemTag and seeker: bootstrapping the semantic web via automated semantic annotation,” in Proceedings of the 12th international conference on World Wide Web (Budapest, Hungary: ACM, 2003), 178-186, doi:10.1145/775152.775178, http://portal.acm.org/citation.cfm?id=775178. 14 “Twine - Organize, Share, Discover Information Around Your Interests | Twine,” http://www.twine.com/. Accessed 1 November 2008 15 David Trastour et al, “A Semantic Web Approach to Service Description for Matchmaking of Services,” http://dating.corante.com/files/semwebmatching.pdf.

16 A. Quan and R. Karger, “How to make a semantic web browser,” in Proceedings of the 13th international conference on World Wide Web (New York, NY, USA: ACM, 2004), 255265, doi:10.1145/988672.988707, http://portal.acm.org/citation.cfm?doid=988672.988707. 17 David Huynh, Stefano Mazzocchi, and David Karger, “Piggy Bank: Experience the Semantic Web inside your web browser,” Web Semantics: Science, Services and Agents on the World Wide Web 5, no. 1 (March 2007): 16-27, doi:10.1016/j.websem.2006.12.002. 18 “Pipes: Rewire the web,” http://pipes.yahoo.com/pipes/.Accessed on 1 December 2008 19 Tim Finin et al, “Social networking on the semantic web,” http://www.emeraldinsight.com/Insight/ViewContentServlet? Filename=Published/EmeraldFullTextArticle/Pdf/1190120503.pdf. 20 Stefan Haustein and Jörg Pleumann, “Is Participation in the Semantic Web Too Difficult?,” in The Semantic Web — ISWC 2002, 2002, 448-453, http://dx.doi.org/10.1007/3-54048005-6_39. 21 Brian McBride, “Four Steps Towards the Widespread Adoption of a Semantic Web,” in The Semantic Web — ISWC 2002, 2002, 419-422, http://dx.doi.org/10.1007/3-540-48005-6_35. 22 Kim H. Veltman, “2002 Challenges for a Semantic Web”,Semantic Web Workshop 2002. Proceedings of the International Workshop on the Semantic Web 2002 (at the Eleventh International World Wide Web Conference), Hawaii, May 7, 2002, Honolulu, Hawaii, pp. 16-22. 23 V.Richard Benjamins et al., “Six Challenges for the Semantic Web,” In KR2002 Semantic Web Workshop 1 (2002): 2004, doi:10.1.1.107.8902. 24 Harith Alani et al., “Towards a Killer App for the Semantic Web,” in The Semantic Web – ISWC 2005, 2005, 829-843, http://dx.doi.org/10.1007/11574620_59.