A knowledge graph is also known as a semantic network and represents a network of real-world entities. In other words represent “reality” as a combination of objects, events, situations, or concepts taking into account the relationship between them. This information is usually stored in a graph database and visualized as a graph structure, prompting the term knowledge “graph.”. The figure below can give you a visual idea of how these entities interlink among each other.

Intuitively, we can note that knowledge graphs take advantages of triples: subject-verb-qualifier in order to infer relationships among the data.

The knowledge graphs is at the core of the semantic web vision and standards like RDF and query languages like SPARQL are maintained by the w3c. A set of pointers for more information is available in this paper that I wrote specifically for interoperability in industry 4.0 and internet of things. However sections 2.3. (of the paper) give an overview of the semantic stack.

Existing tools and technologies that use knowledge graph

A core set of technologies that are currently in productions and used companies like Google, Facebook, Apple includes the following:

• Wikidata: Wikidata is a collaboratively edited multilingual knowledge graph hosted by the Wikimedia Foundation. It is a free and open knowledge base that can be read and edited by both humans and machines.
• schema.org is an organization that publishes documentations and guidelines to using structured data mark-up on web-pages. Its main objective is to standardize HTML tags to be used by webmasters for creating rich results about a certain topic of interest.
• Google Knowledge Graph is a knowledge base used by Google and its services to enhance its search engine’s results with information gathered from a variety of sources.

These technologies are strictly focusing on the data instead and the video below share a practical approach that focus on the data and how to combine them. Note that it does not consider any AI approach but focus on how to combine existing available data and their relationships.

However, it contains very nice examples and give an intuitive idea on how Wikidata, Schema.org and Knowledge graph works.

Research on Knowledge Graph and Artificial Intelligence

Now that we have an idea of what are the basic techniques for representing information we need to understand how we can use machine learning techniques for enriching these information and the expressiveness of the models. What follows is a set of seminars that discuss the knowledge graph from a more theoretical point of view and consider open research questions. This is the first part:

Below you can find the second part. It also contains very nice real life case studies. Including how Siri works!

If you are curious and would like to know more about the topic I encourage you in browsing the YouTube channel for more videos.

On #Knowledge Graph and #ArtificialIntelligence : how to combine data and models for better information. #BigData #DataScience
Share on X

The post On Knowledge Graph and Artificial Intelligence appeared first on Francesco Lelli.