Getting into the Google Knowledge Graph
Searching is changing. After googling something, we no longer see results from documents (web pages); we search over data that comes from the Knowledge Graph. Or without the rhyming, searching is performed over entities instead of text.
For the ones who have never heard about Google’s knowledge graph: a knowledge base that collates millions of pieces of data from a variety of sources across the web to enhance its search results. It is defined as the largest warehouse of knowledge in human history. In this project, Google’s goal is to systematize the way of putting facts, people and places together to create search results that are interconnected, more accurate and useful. This formal definition gives us an insight about the knowledge graph but there may be questions that pop in minds. For me, the questions are:
1- How does google collate data that is meaningful to its search engine(implement it to its knowledge graph)
2- How can we get into the Knowledge graph?
How does google collate data that is meaningful to its search engine(implement it to its knowledge graph)?
Of course, Google can’t do this alone. the company needs us to help make search results valuable and helpful for its users.
Let’s say you are a blogger that posts about features of programming languages and your current blog post is about “bugs on new Java version”. The success of your blog depends on how well you can guide the search engine to interpret that word correctly. Are you referring to the errors on a computer program or “insects”? The word “bug” could mean both of above. if you have a meta description, title, and keywords in your post, google can then sync your content with the appropriate user intention. After all, google promotes your post to the users that probably perform searching using keywords “java” and “bugs”. From the example above, we can easily get that google uses semantic web to build its own knowledge graph. As more user contributes to the graph, it grows and gives us more accurate results.
How can we get into the Knowledge graph?
So far we know how we can be findable in the search results: we must specify the entities of our content. In order to make our content interpretable for Google Knowledge Graph, the first thing to do is to place structured data markup from Schema.org on your web page (or content). Schema.org hosts a collection of structured data markup schemas that helps search engines understand the information contained within HTML documents. This structured data allows search engines to identify entities and define relationship between them which in turn leads to better, richer and more useful search results. In other words, if you are willing to increase the chance of getting into the Knowledge graph, you should mark up your page with the appropriate structured data markup.
Google also leverages information from authoritative resources such as Wikipedia, Google Maps, WikiData and so on. It can be worthwhile to create a page on Wikipedia but you can’t slap a couple of sentences together and post them on Wikipedia; you need some citations which show that your content deserves to be in Wikipedia. So if you have a company with awards or references, it may be more appropriate to create a Wikipedia page to make yourself more visible on Google Knowledge Graph.
Of course, Wikipedia isn’t the only place where the knowledge graph gathers information about the entities. WikiData also exists but you need to be able to back up your claims like Wikipedia. The best way to introduce yourself to Google is to create accounts on social networks and creating an Info Card. This link shows how you can create/edit your Info Card (Knowledge graph card).
Apart from all, Google Knowledge graph seems to be supplanted by
the more sophisticated Knowledge Vault which I have no idea for now but I’m planning to elaborate on and create a new blog post about it :)
