Science | The explanation of the terminology related to knowledge map

The Knowledge Atlas (knowledge Graph) was proposed by Google in 2012. Companies often design new concepts and nouns for commercial purposes. But each concept has its historical origin and essential connotation. Several concepts related to the knowledge map are listed below, and the relationship and difference between them and the knowledge map are briefly expounded.

Knowledge Base: usually translated as " Knowledge Base ". Knowledge Base is one of the classical concepts of artificial intelligence. It was first used as part of the expert system to support reasoning. Knowledge in the knowledge base has many different forms, such as ontology knowledge, relevance knowledge, rule base, case knowledge and so on. Compared with the concept of knowledge base, the knowledge map focuses more on the construction of relevance knowledge, such as ternary group.

The Semantic Web: usually translated as " semantic Web" or " semantic Internet ", is the "1" proposed by the father of the Web, Tim Berners Lee, in 1998. The core connotation of semantic Internet is: The Web is not only to link text pages through hyperlinks, but also to link things up, so that search engines can directly search for things, not just to search the web. Google's knowledge Atlas is the realization of the concept of semantic Internet commercialization. The semantic Internet can also be viewed as a global knowledge base built on the Internet.

Linked data: usually translated as " linked ". Tim Berners Lee was introduced in 2006 to emphasize that the purpose of the semantic Internet is to create links between data, rather than simply publishing structured data online. He developed four principles of "2" for establishing links between data. Conceptually, linked data is closest to the concept of a knowledge map. But the specific implementation of many business knowledge maps does not necessarily follow the four principles proposed by Tim.

Semantic net/semanticnetwork: usually translated as " Semantic Web " or " semantic Web", this translation is often confused with the translation of Semantic Web, in order to differentiate, The translation of "Semantic Web" is used here. The semantic network was first introduced in 1960 by cognitive scientist Allan M. Collins as a method of knowledge representation "3". WordNet is the most typical semantic network. Compared with the knowledge map, the early semantic networks focus more on describing concepts and relationships between concepts, and knowledge maps emphasize the links between data and things.

Ontology: usually translated as " ontology ". The ontology itself is a philosophical term. In the 80 's, AI researchers introduced the concept to the computer field. Tom Gruber defines the ontology as "the formal description of concepts and Relationships", "4". In popular point, ontology is similar to schema in database, it is mainly used to define classes and relationships, as well as class hierarchy and relationship level. Owl is the most commonly used ontology description language. Ontologies are often used to define schemas for a knowledge map.

Citation:

[1] Tim Berners Lee. The Semantic Web Roadmap. Https://www.w3.org/DesignIssues/Semantic.html, 1998.
[2] Tim Berners-lee (2006-07-27). "Linked Data". Design issues. W3c.
[3] Allan M. Collins; M. R. Quillian (1969). "Retrieval time from semantic memory". Journal of verbal learning and verbal behavior. 8 (2): 240–247.
[4] Gruber, T. (1995). "Toward principles for the Design of ontologies used for knowledge sharing". International Journal of Human-computer Studies. (5-6): 907–928.

openkg.cn

The Chinese Open Knowledge Atlas (openkg.cn) aims to promote the openness and interconnection of Chinese Knowledge Atlas data, and promote the popularization and wide application of Knowledge atlas and semantic technology.

Click to read the original and go to openkg Blog