Recently, I am making a tool for the company to realize the import of rdf to knowledge graph. I have heard of rdf before, and I have read it a little bit, but the document of w3c is really obscure and difficult to understand.
It took about a week to read the previous data structure diagrams, concepts such as rdf, owl, semantic web, knowledge graph, etc., and then roughly understand what these do. But it is still not very clear. I started to write rdf analysis last week. During the writing process, I found that it is really useless to look at those concepts. It may be my personal bias.
Personally, I really don't know what rdf is doing just by looking at it. When I analyze the rdf file specifically, I understand it more deeply. RDF attempts to describe resources on the Internet in the form of triples, and use uri to point to these resources. As we all know, each uri can only point to a specific resource. There may be multiple URIs pointing to the same resource, but not one uri points to multiple resources. The so-called triple is similar to the subject, predicate and object in the language. The subject is the resource to be described, the predicate is the attribute of the subject, the object is the attribute value, and the object may be a character or another resource. For example, an article on the Internet may have attributes such as author, title, and publication time, and the author may be a person who has his own encyclopedia. To describe such a relationship, rdf will use these triples to describe these relationships, "article, author, Xiaoming", "article, title, Xiaoming's life in the UK", "article, published time, 2016-010-20" "Xiao Ming, Encyclopedia, Baidu Encyclopedia". RDF stipulates that the resource uses uri, the predicate is also a uri, and the object can be a character or a uri.