Knowledge Graph and the Semantic Web

Have you ever search something on Google? You’ve probably search on Google things like “how many grams are a cup”, “how to convert a video in .avi to .mp4”, “What is the capital of Bhutan” “who wrote Wuthering Heights”. And Google will immediately give you relevant search results according to your queries as Google don’t know the exact answer but you can pinpoint relevant options according to what you have written.

According to the above, Google is a search engine that fulfills its basic objective: to shed search results that can meet needs like a good converter measures or formats, a good encyclopedia, translator, etc. For a long time what Google has done is recommended, but this time I want to show that with the evolution of the Internet, the amount of content available in the cloud and with the possibilities of optimization, Google will not only recommend, but it is also working to give answer instead of just the search results.

The Semantic Web

The web is built from HTML documents, tag-based markup language that that organizes the content of a site to display the titles as titles, organized into paragraphs, tables and text, and audio-visual objects we want to insert. Labels act as instructions for the browser.

However, the labels not only serve to give instructions to display, but also allow you to add additional information about our content, which we will only read by the browser and search engines. This means that it is possible to describe what we have produced in the eyes of Google to be considered not only as a product but as something more specific like: a review of a book or a movie, a product with all its details, a recipe, contact information, and more.

The Web is full of written content, so the semantic web is a set of activities that are based on the idea that in order to improve user experience on the Internet it is possible to add meanings to such content to describe, understand and relate them.

Here’s an example. When we write something for the web, a browser would read a title like this:

The title is fine as it is written. But if you want to contribute to the semantic web you should specify the element by adding an attribute that says that “Kill Bill Vol.1” is a film. This would make all the data I am posting are taken into account so that Google has an increasingly clear idea of “Kill Bill Vol.1”. So, next time someone write “Kill Bill” in the search bar, Google will not only just recommend search results but it will also have the answer right there.

Knowledge graph and the semantic web

In 2012, Google added the search engine chart of knowledge based on semantic search to provide structured responses to users, along with search results of other websites.

Tim Berners-Lee was the creator of HTML language, the object locator system URL, and HTTP protocol and most of the web is based on those technologies. His initial idea was directed towards the semantic web. So in 1994 he founded the W3C, an international community that works to develop guidelines and standard guidelines, such as schema.org, in order to make the best potential of the web. The W3C aims to support a web-based data to enable computers to do a better job in terms of our experience on Internet.

Components of the Semantic Web

In principle, the semantic web is composed of meta and frameworks that enable interoperability of machines such as XML, RDF and SPARQL.

XML is a markup language like HTML, which stores data in structured documents; RDF is a framework that defines the conceptual model to describe data; SPARQL is a query language for RDF. Those three are interrelated and essential and basic to the development of semantic web technologies.

Currently, these components have been condensed into a single language called OWL ontological marking, which is designed for applications that need to process information in addition to filing for internet users. OWL is part of the stack of W3C recommendations related to the Semantic Web. It is a language that goes beyond XML and RDF, as it provides more vocabulary for describing properties and classes.

Barriers semantic web

By understanding the semantic web as a concept three specific barriers that make this not a complete reality on the web today arise.

1. The way to contribute to the Semantic Web lies on one thing: data. For Google to take data from different web sites is necessary that all contents are written in a standard way, considering the above components. Only then, the browser and the search engine can identify the relationship between them.
   To achieve this it is necessary that all webmasters we edit the content of our sites manually, according to the guidelines of the W3C , and this is an expensive task not everyone is willing to make, considering that the end is a common good.
2. The Semantic Web is a clear threat to living websites advertising, because the consequence that the contents of the web are written in a standard way is that Google could provide answers to user queries in their search results page . In that sense, it becomes unnecessary to visit the sites listed in the SERPs, regardless of their position, and this would be another reason not to optimize contents for the semantic web.
3. Many of the people working on the Internet, especially in the production of content. The idea that Google has strayed from its original essence of the filter and to rank the best content for users, and that’s one of the main reasons why most people do not consider the optimization of content. A clear example of this is the misconception that Google tries to give competition to other social networks Google+ when it actually seeks to eliminate the anonymity of the web. Associating our content with our Google+ accounts was a very good SEO practice until 2014. As this not only contribute to our authority and online reputation but gave us ranking points because we were working with the description of the content available on the web.