The Use of Multilayered Extended Semantic Networks for Meaning Representation

The framework of Multilayered Extended Semantic Networks (abbreviated: MultiNet) is one of the few knowledge representation paradigms along the line of Semantic Networks (abbreviated: SN) having a comprehensive, systematic, and publicly available documentation. The paper describes the main features of MultiNet and the standard repertoire of representational means provided by this system, the application of MultiNet for the meaning representation of natural language expressions, and the software tools connected with it. Besides of the structural information, which is manifested in the relational and functional connections between nodes of the SN, the conceptual representatives of MultiNet are characterized by embedding the nodes of the network into a multidimensional space of layer attributes. To warrant cognitive adequacy and universality of the knowledge representation, every node of the SN uniquely represents a concept, while the relations between them have to be expressed by a predefined set of about 110 semantic primitive relations and functions, which are described on a metalevel by means of an axiomatic system of second order predicate calculus formulae. The classification of MultiNet nodes into sorts (which form a conceptual ontology) is an important basis for the definition of the domains and value restrictions of the above-mentioned relations and functions. To support the effective work with MultiNet and to deal with the semantic phenomena of natural language, MultiNet has been provided with several software tools, which include: MWR: A workbench for the knowledge engineer supporting the graphical representation and manipulation of MultiNet networks as well as the accumulation and management of MultiNet knowledge bases. NatLink: An interpreter which automatically translates natural language sentences into MultiNet semantic networks by means of a word-class controlled syntactic-semantic analysis. LIA: An interactive workbench for the computer lexicographer which is used to create large semantically oriented computer lexica based on the expressional means of MultiNet. All three tools are important prerequisites for the use of MultiNet as a semantic interlingua in NLP systems.