Semantic Web Applications

The reason for the enhancement of the Web, which eventually has led to the emergence of Semantic Web, is so that it can be of use to our day-to-day lives. This chapter takes a deeper look into the various applications of Semantic Web including e-commerce, e-business, knowledge management, search engines, Internet of Things, and Semantic Web services.

E-Commerce

E-commerce has become a phenomenon in the world of business; creating immense business opportunities where both large-scale and small-scale online retailers can compete on a level ground. The main features of e-commerce are online transaction processing and electronic fund transfer. But apart from these two, there are many other functions involved including newsletter sign-up and products book-marking. Thanks to e-commerce, online users are able to view a wide range of products and make product comparisons on different grounds like features, prices, and warranty (Holak & Cole, 2018). Thus, a simplified consumer-producer relationship is established, whereby the consumer makes an easy search for the product online and orders it from the right producer. With e-commerce, geographic limitation is eradicated, opening ways for the purchase of a product from any corner of the world. That is why the term ‘global village’ is commonly used (Ecommerce Gudie, 2018).

Semantic Web offers immense opportunities for e-commerce to thrive. Its Ontology languages act as a bridge between Client-side language RDF and user-side language SPARQL, offering one of the most versatile solutions for e-commerce(Ercim-news, 2009). There are three key Semantic Web e-commerce levels, namely: Semantic Web services level, Service distribution level, and Real-time Input/output level(Enoh, 2013).

  • Real-time input/output level: This level is exposed to both the client and end-user. The clients are given an opportunity to publish their product details as an input which is then fed into the Semantic web form, while end users search for specific products using certain keywords.
  • Service distribution level: This stage involves mainly Agents, consisting of pre-defined and well-defined task module whose duty is to gather information and process tasks. Both the client and end-user have access to the Agent. The client-side agent gathers the information and passes to the relevant query language in RDF. On the other hand, the end-user Agent has the duty of receiving this information and extracting the exact information needed by the user. In order to ensure accuracy, it has information-validating features in reference to the keywords supplied by end-user.
  • Semantic Web services level: Has query languages which actively exchange information regarding various operations through ontology. The client-side retrieved information is parsed by the Agent to RDF after which another Ontology performs data linking. At this juncture end-user side Agents communicate with each other to gain access to relevant information as requested by the product searcher.

Prior to Semantic Web, end-users were only served by centralized e-commerce platforms when comparing prices of different dealers of a given product. The suggestion of product offers was solely dependent on dealers or suppliers publishing such a product. This meant that there was a limitation on the percentage of products placed on offer. Semantic Web overcomes this challenge by extracting product details directly from the website of the given dealer. In the long run, consumers get a holistic view of the entire market since consumers do not just see products that are on offer on centralized platforms, but all the products on offer across the Web (Heidari, 2009).

E-Business

We live in a world where without access to knowledge and information, business and economic performance are greatly jeopardized, as sustainable development becomes an almost-impossible engagement (Bedini, Gardarin, & Nguyen, 2011). E-business is an approach through which business goals are achieved, aided by technology which facilitates information exchange across value chains, as well as technology which brings about better decision-making (Bedini, Gardarin, & Nguyen, 2011). One of the most effective ways that organizations can improve their e-business processes is through inter-organizational collaboration which enhances value propositions. Businesses operate in a dynamic environment and act as knowledge-based organizations.

Semantic e-business is an approach through which knowledge is managed so as to facilitate the coordination of e-business processes via the application of Semantic Web technologies in a systematic manner. Key progress made in Semantic Web technologies like knowledge representation, Web services, multi-agent systems, and ontologies are borrowed to make Semantic e-business a success. These technologies bring about some level of transparency, facilitating free exchange of semantically enriched knowledge and information such as know-how and content (Buraga & Cioca, 2011), as well as facilitating collaboration among various organizations.

There are some key characteristics which today play a crucial role when it comes to e-business. These include: globalization, networking, and collective intelligence (Cardoso, Lytras, & Hepp, 2014). Globalization entails knowledge and information being created and consumed from the perspective of the global world. With this kind of view, it is essential that local boundaries be eliminated and that organizations be willing to export what they are capable of doing beyond the confines of local boundaries, aided by mechanisms which permit the understanding of opportunities and threats. Concerning networking, we live in an era where business and economic activities, in addition to competition, call for new models that facilitate business networking. This view holds that social and business networking at the global level is the essence of emerging demands within the business world (Cardoso, Lytras, & Hepp, 2014). Collective intelligence as globalization, leads to a state where business networking among organizations happens on an open platform, it becomes essential to apply intelligence filters in the global information world. This kind of trend renders traditional profitability, marketing, and performance business models obsolete, hence the need for a new and dynamic approach – Semantic Web (Singh, Iyer, & Salam, 2005).

Semantic Web technologies are the backbone of e-business, providing an open paradigm which fuels success of the latter. Open paradigm has to do with contemporary movements such as open knowledge, open access, open culture, open content, open source software, and open research. Such an approach goes a long way in supporting new business models(Cardoso, Lytras, & Hepp, 2014).

Knowledge Management

Large organizations always face a challenge when it comes to knowledge management. There are different forms in which knowledge resides including tacit knowledge in individuals and as explicit knowledge in documents (Huang, 2012). There are also extreme cases which lies between the two. The best way an organization can be able to become better at knowledge management is by placing key attention on strategically important knowledge. There are different ways in which knowledge management can be viewed including: acquisition, utilization, storage, creation, and validation.

The main goal behind the development of Semantic Web is to ensure availability of intelligent services through provision of machine-readable content. The success of Semantic Web largely relies on ontologies. Thanks to Ontologies, there is a common domain through which communication between people and applications happens. This is an important aspect for knowledge management as well. Ontology has a specified vocabulary which facilitates information exchange (Back, Vainkainen, & Juhola, 2014). There are five key components in ontologies which formalize knowledge, these are: General/common ontologies, task ontologies, domain ontologies, meta-ontologies, and knowledge representation ontologies. Each of these have features which are helpful to knowledge management.

But the most impactful feature on Semantic Web is the ability to have more effective searches. For example, a firm may decide to organize its content on the basis of an ontology, which could be exploited for the improvement of the quality of searches. Knowledge management is normally faced by the challenge of finding relevant information, therefore, ontologies are quite helpful here (Livesey, 2013). In particular, the management of multi-media content would be simplified with the aid of ontologies. The different images, videos, or audio could be semantically annotated based on ontologies. Indexing and searches are supported by these annotations. Given that individuals tend to have their own description for videos, audios, images, and other non-textual objects, it becomes crucial that searches are able to go beyond the limitations of keywords.

The creators of Semantic Web envisioned a situation where applications were responsible for most of the hard work whereas humans only dealt with tasks which needed decision-making or controlling of the application (Keston, 2012). It is like to a service paradigm from tool paradigm. The achievement of this vision requires that machines are made as intelligent as possible, so that they can be able to do things like scheduling meetings at the most convenient time, aided by ontologies which penetrate deeper levels for knowledge. Thanks to Semantic Web, knowledge management is being simplified in such a way that organizations or companies could not have envisioned before, adding value to the available knowledge. It would be of no sense to have vast amounts of knowledge that cannot be exploited judiciously.

Search Engines

The last few years have recorded a skyrocketing amount of data on Semantic Web, much of which is available on the Web and can be easily exploited. As opposed to the classic Web, Semantic Web offers a friendly interface powered by software programs. In order to meet the needs of end-users, it is vital that there be infrastructures that can help find and select data that is distributed throughout the Semantic Web. Developers understand this, and have ended up creating some of the most powerful Semantic Web search engines known to mankind.

The design principle for these systems is not the same as the other search engines as we know them, and they have a unique support for applications and user needs (Domingue, 2011).

As at the moment, Google, Bing and Yahoo still dominate the world of search but with regard to Semantic Web, there are more performant search engines have emerged. A good search engine is one that is able to match queries with the context and then return results based on that context. The following are some of the emergent Semantic Web Search engines (Mankani, 2016):

  • SenseBot: The technology behind this search engine is able to summarize top results after a given query, going past the need to get deeper into URLs to extract information that is being sought by the end-user. There are a wide range of other services that have been built around this search engine including Link Sensor which can automatically select key concepts out of a blog post and then link it directly with other articles by the blog or publisher (Radhakrishnan, 2007).
  • Cognition Search: This Semantic Web search engine has the ability to extract key results out of a given content, including fetching relevant ads. The technology can be accessed via APIs created by the company.
  • Exalead: Designed specifically to search images, this Semantic Web search engine is one of a kind as it allows users to narrow down their searches based on some factors like image size, color, as well as content.

Most search engines have gone ahead to copy some of these features, indicating their versatility and usefulness. As at the moment, Exalead is being designed with special attention on enterprise search market (Jmornini, 2011).

  • WalmartLabs: Previously known as Kosmix, the search engine has the blessing of Walmart, hence holding a higher potential to breakthrough the market. The idea of search here is taken a notch higher by including a dashboard whose purpose is to guide users around the Semantic Web.

The fact that it concentrates on informational search makes it a great choice for persons who would like information on specific topics and not just get URLs or a particular answer (Wikipedia, n.d.).

Internet of Things IoT

The Internet of Things is fast becoming a reality. When fully developed we shall have hosts of connected devices talking to each other. Much of the progress made in IoT is attributed to the breakthroughs that researchers have had in Artificial Intelligence and Deep Learning. But that is not enough to make Internet of Things a reality. Some semantic backbone is essential if the technology has to excel.

It is estimated that by 2020, there will be over 50 billion devices that are connected, paving the way for swift communication amongst these devices (Dataversity, n.d.). But that communication will only be possible if these devices are set to standards understandable to each one of them. And that is where Semantic Web comes in.

The increasing interest on the Internet of Things is fueled by the desire to develop high-quality applications which seamlessly satisfy consumer needs. From e-heath, smart buildings, to smart cities, the applications of IoT and web technologies are limitless and with the certainty that this will continue to increase. The future for IoT and web technologies is brighter. At the same time, semantic technologies applied in a wide range of domains have proven that they are more effective at meeting specific user and organization needs. Semantic Web technologies have been able to do, among other things: (1) deduce new knowledge required for the construction of smart solutions, (2) aid in the integration of data application, (3) provide semantic interoperability, thereby mitigating heterogeneity, and (4) aid in the interoperability of different data processes like storage, management, and representation of data (Dataversity, n.d.).

Semantic Web, Web technologies, and IoT technologies could be combined, leading to the evolution from Internet of Things to Web of Things and eventually to Semantic Web of Things. At the level of Semantic Web of Things, it is assumed that machines would be able to communicate in the most open manner, freely exchanging key linked data, thus making it possible for the initial objectives of IoT to be met in the most effective way.

There exists various open source tools that have been designed with WoT, IoT, and Semantic Web of Things in mind. M3 (Machine-to-Machine Measurement) is one of such tools. This is a framework that is able to interpret IoT data, enabling the design of interoperable cross-domain or domain-specific SWoT applications(Sensormeasurement, 2015). M3’s main goal is to make it possible for developers to quickly prototype IoT applications, thereby reducing the learning curve when integrated with Semantic Web technologies. The main issues addressed by the framework include ontologies, interoperability of sensor data, and use of the data to create WoT applications.

As a result of Semantic Web, key ideas and technologies are presented which are required for solving difficulties presented by the current state of IoT. Knowledge-based systems are able to freely access information sources, providing the transparency needed for connected devices. As at the moment, realizing IoT has remained a challenge because this transparency is blurred, but Semantic Web of Things promises to change the stalemate.

Semantic Web Services

A Web service refers to website that goes beyond the supply of static information, making it possible for the execution of certain actions (services) in an automatic manner, for instance when selling a product or controlling a physical device. In order for this to become a reality, Web services rely on XML standards such as a messaging protocol, a description protocol, WSDL, and SOAP, all backed up by some kind of semantic expressiveness. For instance, WSDL may be able to provide a description of the interface of various services and the deployment of these services via SOAP but it remains limited when it comes to expressing the overall competence of particular services. Semantic Web Services on the other hand overcomes such kind of limitation, specifying the interfaces as well as providing a full description of the capabilities and prerequisites for use due to the OWL-based ontologies (Vagra & Hajnal, 2005).

The Semantic Web service technology is a convergence of service-oriented computing and Semantic Web. Through the use of Intelligent Agents software, the challenges of interoperable, meaningful coordination and automated Web Services are addressed. There are some key factors which can be taken into consideration for each semantic service description framework, namely: (1) the type of service semantics that are described, (2) the language or formalism involved, and (3) the reasoning level that brings about abstract service descriptions (Terziyan & Kononenko, 2013). Generally, a given service’s functionality is best described according to what it does and the manner in which it works. The service process model is used to capture its functional semantics. With this profile, we are able to get the service’s signature, paying attention to the input (I) and output (O) parameters, as well as the preconditions (P) and effects (E) which ought to be true once the service is executed in the world or state or prior to the execution (Terziyan & Kononenko, 2013).

Progress in Semantic Web services has made it possible to deal with issues associated with composition, execution, and automated service discovery, among other problems linked to Web services standards, fueled by additional semantic layers. Thanks to the extra semantically transparent layers, clients gain the ability of successfully using services which are discovered in a dynamic manner without the need for the client and service developers to have had some kind of negotiations.

The need for formally grounded descriptions when it comes to Web Services is supported by the reasoning that agents must be able to have an understanding of both functional and non-functional semantics based on proper logic reasoning. As a result, an assumption is normally made that; a given logic reasoning is often in agreement with some kind of semantic service description model (Vagra & Hajnal, 2005).

Leave a Reply

Your email address will not be published. Required fields are marked *