A Brief History of the Semantic Web
Imagine looking for an ophthalmologist. The search engine not only gives you a name but also shows you which doctor is the best for you and when he has free appointments. The Semantic Web could make such and much more complex search queries possible – because it catalogs and organizes human knowledge that is available worldwide.
What is the Semantic Web ?
The Semantic Web or Web 3.0 relates information on the Internet to one another. Search engines can not only search for keywords but also evaluate them and extract meaning from them. The semantic web, among other things, enables more intelligent tools to be developed.
Semantic Web: Not a new idea
The idea of developing a smarter, more meaningful, the semantic web has been around since 2001. That year, Internet inventor Tim Berners-Lee published his article ” The Semantic Web ” in Scientific American, kicking off Web 3.0.
Texts mean something to people. They understand synonyms and contexts, they have associations and ideas. It is different with computers: they perceive text but do not understand it. In his article, Berners-Lee called for the World Wide Web to be enriched with meaning, i.e. to make it semantic. He dreamed of so-called search agents, which should be a lot more powerful than search engines as we know them today.
As an example, Berners-Lee cited a woman who wants to make a doctor’s appointment. The search agent would research which doctor is based in their area, what specifications the individual doctors have, and how other patients rate them. Then he would point the woman to the doctor he believed would best suit her needs. He would also match her schedule with the doctor’s and allow her to make an appointment immediately.
With a simple search query, a whole series of steps would always be carried out. The internet would become smarter and could relieve people of a lot of time-consuming work.
How does the semantic web work?
Search agents cannot understand connections on their own. That is why people have to help and enrich the data on the Internet with meaning so that computers can capture them. For this purpose, texts are broken down into so-called triples using the Resource Description Framework (RDF). These usually consist of subject, predicate, and object.
Tim Berners-Lee (subject) – is the inventor (predicate) – of the World Wide Web (object).
RDF works across websites. This means that search agents can search for the individual parameters on all prepared websites. This makes it possible, for example, to display all publications by a certain author in an orderly manner or to search recipes specifically for which ingredients they contain.
Ontologies create connections
So-called ontologies create – roughly speaking – connections between the web content. For example, search agents can understand that a word can have several meanings and is used in different contexts. If there is an RDF with the subject “Angela Merkel”, the search agents know that said person is not only “Federal Chancellor of”, but also “Boss of”, “Wife of”, “Daughter of” and “Customer of”.
Ontologies ensure that terms are not simply classified as synonyms, but are classified more profoundly. At the same time, it also helps search agents understand that a term does not only apply to one person or thing. Not every person dressed in orange works in the city cleaning department, not every student gets a scholarship and not everyone with a driver’s license is allowed to drive a tram.
Web 3.0 is coming – but slowly
The Semantic Web opens up completely new possibilities. In theory, you could easily schedule a conference by automatically comparing the calendars of all the participants you want. If you would like to write an academic paper, you could display a list of relevant literature, sorted by relevance. And if you want to invite someone to a business lunch, you could search specifically for restaurants that offer your favorite dishes, are not fully booked at the time of your choice, and have a maximum of ten tables. All of this with a single search query.
There are already first attempts to put the semantic web into practice. For example, Google practices with recipes. For some recipes, the ingredients and other information (such as the fat content) are displayed in addition to the pure search results and users can refine their search query based on this information. However, the operators of the websites on which the recipe appears must manually provide Google with the information in RDF format.
The Google update “BERT” is also about better understanding human language. The update should enable Google, among other things, to record the relationship between individual words. In this way, the search provider wants to be able to react even better to the search intention.
The search engine Semager shows words that are related to a search term. “Dog” also includes “hut”, “leash” and “food”. The search engines and other tools have not yet reached the options described above. It will probably be many years before everyone can use Web 3.0 as a matter of course.
Semantic Web will change the search engines: The semantic web will enable people and computers to work together better. This offers an almost infinite number of ways to simplify work processes and save time and resources in all areas of life. What search engines like Google are doing today may one day seem laughable to us. But there is still a lot to do before then.