Semantic networks are a form of knowledge representation.
“A semantic network or net is a graphic notation for representing knowledge in patterns of interconnected nodes and arcs. Computer implementations of semantic networks were first developed for artificial intelligence and machine translation, but earlier versions have long been used in philosophy, psychology, and linguistics.” (Sowa, retrieved 19:33, 14 August 2007 (MEST).
“Knowledge representation is an issue that arises in both cognitive science and artificial intelligence. In cognitive science it is concerned with how people store and process information. In artificial intelligence (AI) the primary aim is to store knowledge so that programs can process it and achieve the verisimilitude of human intelligence.” (Wikipedia, retrieved 19:33, 14 August 2007 (MEST)).
1.1 In cognitive theory
For some authors knowledge is stored either in episodic or semantic memory. The further is organized in spacio-temporal dimensions, the second according semantic content-oriented principles, e.g. networks of concepts.
“Long-Term Memory which is a large storage system, stores factual information, procedural rules of behavior, experiential knowledge , in fact everything we know. We have two types of long term memory Episodic and semantic memory Episodic memory represents our memory of events and experiences in a serial form. It is from this memory that we can reconstruct the actual events that took place at a given point in our lives. The second is Semantic memory, which is a structured record of facts, concepts and skills that we have acquired . The information in semantic memory is derived from that in our episodic memory, such that we can learn new facts or concepts from our experiences. Semantic memory is structured in some way to allow access to information representation of relationtionship between pieces of information and inference. One model for the way in which semantic memory is structured is as a network. Items are associated to each other in classes and may inherit attributes from parent classes. This model is known as a Semantic network.”(Cognitive Psychology on Memory, retrieved 20:17, 14 August 2007 (MEST)])
1.2 In computer science
In computer science, a semantic network can be defined as a knowledge representation formalism which describes objects and their relationships in terms of a network consisting of labelled arcs and nodes.
- “A semantic network is often used as a form of knowledge representation. It is a directed graph consisting of vertices which represent concepts and edges which represent semantic relations between the concepts.” (Wikipedia, retrieved 19:33, 14 August 2007 (MEST)).
- “A semantic network is a knowledge representation tool consisting of a framework of semantically related terms, with the purpose of allowing a definition of those words through their relationships.” (, retrieved 19:33, 14 August 2007 (MEST)).
See also: Ontology (most ontologies use a kind of semantic network).
2 Cognitive semantic networks
“The most prevalent example of the semantic network processing approach is the Collins & Quillian Semantic Network Model. The semantic network processing approach states that the meanings of words are embedded in networks of other meanings. Knowledge is validated and acquires meaning through correlation with other knowledge, (Harley, 1995). The connections within a semantic network are not only associative in nature. The links between information in a semantic network are qualitative and purposeful. Therefore, the links within the network have semantic value.” (Wikipedia, retrieved 20:17, 14 August 2007 (MEST))
In the Collins and Quillian (69) model semantic nets are composed of simple concepts, concrete-abstract (is-a, ako) relations and part-whole (attribute, is, has, can) relations. E.g.
Parts of the knowledge about the concept Ms. X, member of parliament could be represented as follows:
human ^ | | politician --is--> in party y | ^ is-a | | is-a | | | | Ms. X ----is--------> MP ---can---> vote in p.
Below, we quote from the Wikipedia article (copyright: GDFL!):
In the Collins & Quillian model, concepts are represented as nodes that are interconnected to other nodes within the network. The nodes are accessed when they are heard and then activated in memory causing information that is correlated to the concept to be primed. The ISA link is the most common link in this semantic network model. The nodes within the network that are connected by this link have a specific type of relationship that is hierarchical in nature. Therefore the concept at the lower level node is a form/type of the concept at the higher level node. This type of interconnectedness is also evidenced by the HAS A link.
The structure of the Collins & Quillian model compensates for many of the deficits identified in earlier theories such as the behaviorist notion of meaning being derived from a network of associations. According to the Behaviorists, a word is defined based on placement in a network of associations. Meaning is extrapolated from an accumulation of episodic instances involving the word in question. The primary problem with this theory is the fact that a definition based solely on associative properties alone is inadequate since it does not encompass all aspects of meaning, (Harley, 1995). Other prominent flaws in the network of associations include its lack of structure, failure to evince relationships between words and the lack of cognitive economy.Several components of the Collins & Quillian semantic network model address the issues identified in the Behaviorists' network of associations. The hierarchical schema intrinsic to the semantic network model creates structure and facilitates the use of cognitive economy. Rosch (1999) refers to cognitive economy as the means by which an organism acquires a substantial amount of information without having to undergo a search of all finite resources. The hierarchical structure of the semantic network model eliminates redundancy since access to information stored at one level is not required to process an instance of the category at another level. For example, if fur is stored at the level of dog it is not necessary to process Collie which is a type of dog. Moreover, the relationships between words are demonstrated by the purposeful interconnectedness of linked activated nodes within the network.
2.2 Rummelhart and Ortony
The schema theory of Rumelhart and Ortony (1977) claims that personal knowledge is stored in information packets or schemas that comprise our mental constructs for ideas. Each schema we construct represents a mini-framework in which to interrelate elements or attributes of information about a topic into a single conceptual unit. These mini-frameworks are organized by the individual into a larger network of interrelated constructs known as a semantic network. These networks are composed of nodes: representations of schemas. Ordered labelled relationships define the propositional relationship between the nodes. Chris Muller, retrieved 19:33, 14 August 2007 (MEST).
2.3 In computing
See the Wikipedia] article for now ....
3 In education
The hypothesis for educational technologists is that mapping the semantic network of an expert or knowledgeable person onto the structure of a hypertext and then exposing the learner's to it will contribute to the development of the learner's knowledge structures
Research shows that semantic networks are somewhat transferable: as a result of instruction, learners' knowledge structures more closely resemble the instructor's knowledge structure. So, learners are acquiring two things during instruction:
- isolated knowledge
- knowledge structures that mimic the teacher's knowledge structure
See also: Educational use of concept maps and concept learning
- John F. Sowa's website (inventor of conceptual graphs).
- Cognitive Psychology on Memory
- Collins & Quillian Semantic Network Model (Wikipedia)
- ConceptNet (quote) “is a semantic network containing lots of things computers should know about the world, especially when understanding text written by people. It is built from nodes representing words or short phrases of natural language, and labeled relationships between them.”
- Harley, T. A. (1995). Semantics. Chapter 6 of The psychology of language: From data to theory. East Sussex, UK: Psychology Press, pp. 175-205.
- Rosch, E. (1999). Principles of Categorization. In E. Margolis & S. Laurence (Eds.), Concepts: Core Readings (pp. 189-206). Cambridge, MA: MIT Press. Reprinted from (1978), E. Rosch & B. Lloyds (Eds.), Cognition and Categorization. Hillsdale, NJ: Laurence Erlbaum Associates
- Jonassen, David H. and Sherwood Wang (1993), Acquiring Structural Knowledge from Semantically Structured Hypertext. Journal of Computer-Based Instruction, Winter 1993, Vol. 20, No. 1, 1-8.
- Maida, Anthony S., & Stuart C. Shapiro (1982) "Intensional concepts in propositional semantic networks," Cognitive Science 6:4, 291-330.
- Sowa, John, F. Semantic Networks, Rrevised and extended version of an article that was originally written for the Encyclopedia of Artificial Intelligence, HTML, retrieved 19:33, 14 August 2007 (MEST).