The Semantic Web is presented and led by the World Wide Web Consortium W3C. In the conceptual web, website owners are encouraged to include semantic content in their web pages to make them more functional and shareable. The conceptual web, in a simple and comprehensible definition, is the future of the current web in which all parts can be understood and processed by machines in addition to humans.
The question is, why should machines also be able to understand or process the information we need? The answer is very simple to help us humans, more. For example, information is somehow recorded and called by machines so that when you search how to change the language of Windows 10 on the Internet, the search engine will show you the answer instead of displaying tens of thousands of links as an answer.
Although part of this path has already been traveled, it is still far from the expected future. For a better understanding of the semantic web, in the image below, the phrase how to change the language in Windows 10 is searched in English.
History of the Semantic Web
The concept and model of the semantic network were invented in the early sixties by scholars such as Collins, Jacqueline, and Elizabeth F. Loftus, famous linguists and psychologists, in various publications to display structured knowledge conceptually. But it was invented and presented in a specialized way in the field of the web by Tim Bernersley, who is also the inventor of the World Wide Web.
Tim Bernersley, head of the World Wide Web Consortium (W3C), believes that the web, like humans, reaches its flowering stage after the age of 20, and the web’s capabilities are still underutilized. Currently, he supervises the development of the required and proposed standards of the conceptual web. However, it should be noted that in many cases, the technologies proposed by the W3C existed in another form before the W3C developed them.
The importance and current need for conceptual web
Every day, millions of users search for thousands of requests for specific information on the Internet. Fast and easy access to the desired information is possible through the Internet without geographical restrictions. But these widely used services also have disadvantages.
For example, publishing unreal content, not specifying the best and most correct answer, displaying the answer in exchange for introducing thousands of ways to reach the result, and wishing to use an intelligent web that acts like a human assistant caused the feeling of the need for technologies that From a semantic point of view, understanding the request to receive information and making the right and desired information quickly available to users, or providing the necessary services to implement requests remotely, should be on the agenda of the relevant people.
Structure of the Semantic Web
Conceptual web structure is based on a correct understanding of how the user searches and the information on the web. That is, when a user searches for a phrase, the search engine, like a human, must understand the feeling in that phrase. For example, recognize that this phrase is a question, scientific, entertainment, or news. Then, based on that, display the appropriate content to the user, which is a response to the feeling of the sentence.
Therefore, the expected semantic web should master linguistics and philosophy to analyze and categorize all web contents, including pages, images, and other web elements. For example, a search engine indexer should understand a sentence’s subject, object, and predicate and recognize the continuity of sentences in a paragraph when reading a text.
Therefore, the structure of the semantic web is such that instead of collecting discrete data that has no science, it stores and displays understandable information. For this reason, it was necessary to change the display language in websites or HTML and go from the visual aspect, which was only based on how to display content for human understanding, to a concept for robots and all kinds of machines. For example, instead of the b tag, use the semantic tag strong to bold important content so that the robot for indexing can easily understand that this part of the text is of great importance in the subject of the article.
In fact, before the semantic web and HTML5, crawlers or search engine robots only indexed several keywords and phrases and displayed what they had collected to the user based on user searches. But the expectation of today’s society is not fulfilled by showing these small things.
One of the goals of the Semantic Web is to act like a full-fledged assistant and to be able to delegate certain tasks to it. For example, when we want to see a doctor, he can find an appointment at the nearest place and time with the most suitable price from a well-known doctor on the relevant websites and return the result to confirm the registration.
In this situation, we only tell the corresponding robot which doctor’s visit we want in which time frame. To achieve this goal, a special architecture has been established, which has a special layering that cannot be achieved overnight. Therefore, based on a semantic web philosophy, it has several layers and sections, the most important of which include the XML layer, the RDF or Resource Description Framework layer, the ontology layer, the logic and proof layer, and finally, the Trust layer.
1) XML layer
It is better to get acquainted with the concept of XML first. XML stands for Extensible Markup Language. Here we are going to explain a little about its most important part, Markup. Markup is based on metadata. Metadata is information about information and is separate from the information itself; when we want to tell the browser that a sentence is a title and the subject of our article, we specify it using a sign.
HTML is also one of the classic markup languages. For example, when we specify the article’s topic with the H1 tag and tell the browsers how to display this information, it tells the indexing robots that this section is the article’s topic. That is the most important part of the article! This makes the content more understandable for search engine bots.
Problems in building the Semantic Web (XML instead of HTML).
2) RDF layer
As mentioned, metadata is data about data itself. For example, consider the shelves of a book store, which refer to the type of content of the books on those shelves through cartables. Those cartables are the future data or metadata that provide us with information about the books, which are the same data.
RDF acts as the author of these cartables and considers all web objects, including texts, images, files, etc., as resources that can be used to express statements. Then RDF starts to build metadata for certain attributes according to the general structure of these sentences.
In fact, RDF was designed to provide a general way to define information in a way that can be read and understood by computer programs. RDF descriptions and documentation are written in XML. The XML language in which RDF is used is called RDF/XML. Also, RDF definitions are not displayed on the web. RDF identifies everything using web identifiers, or URIs and defines resources with attributes and attribute values.
RDF Use Cases
RDF takes the lead in describing all parts and elements in the semantic web. For example, description of features for commercial goods, such as price and available quantity, description of the information about articles and web pages such as subject, author, date of creation and modification, description of web images, etc., are among the tasks performed by RDF.
3) The layer of ontology or ontology
Ontology is the philosophical study of the nature of existence, existence, or reality. It is part of the main branch of philosophy known as metaphysics. Ontology deals with questions about classifying things according to similarities and differences.
But there is a fundamental difference between ontology in philosophy and ontology in computer science, and especially ontology in the semantic web. Philosophy reaches ontology through the order between concepts, but computer science does not have such an order and derives ontology from the order we consider for concepts. Ontology in philosophy has a comprehensive view and tries to examine all the concepts. In contrast, ontology in computer science has a much smaller scope, and only the things in the area we need are discussed, which makes the work up to about impress a lot.
But the ontology in the semantic web has a narrower view of the concepts, and we should use it only to solve and overcome the problems in RDF. To put it more simply, we try to deal with equivalent concepts in the semantic web ontology. Therefore, it can be said that ontology has a logical position.
4) Logic and proof layer
As mentioned, the purpose of the semantic web is the joint operation between humans and computers. Accordingly, a human must write content that the machine can understand, and the machine must return content in the human question that can be understood by him, which means a conversation between a human and a machine. But the possibility of conversation between humans and computers will not mean the completeness of the web.
Machines must use languages to describe ontologies that allow conversion to an appropriate level of descriptive logic. Descriptive logic is very suitable for the structure of the semantic web because it is one of the languages of knowledge representation used for the verbal and structured representation of knowledge. Languages such as OWL, RACER, and Kand L-ONE, which are based on descriptive logic, are used in the semantic web.
5) Trust layer (TRUST)
In the last layer, i.e., the trust layer, the quality of web information and its reliability are measured. Currently, on the web, this is not possible for users, and only through some facilities such as SSL and some symbols and protocols do they get this assurance to some extent. But does this assurance also apply to the content published on the web, or can dorsal people use the existing methods? Is privacy possible for everyone?
For example, when you buy from a seller on social networks like Instagram, can you trust him? For this reason, the topic of trust has become one of the current web problems. However, in the Semantic Web guidelines, building trust to eestablishingsafe interactions is a priority, and if it is implemented, the Semantic Web can work very well.