MultiNet

MultiNet is a formalism for knowledge representation and semantic representation of natural language expressions. The acronym stands for multilayered extended semantic networks ( multi-layer extended semantic networks).

The MultiNet paradigm was developed by Prof. Dr. Hermann Helbig as an extension of earlier semantic networks. It is a formalism which aims particularly in contrast to other methods of knowledge representation on the presentation of natural language formulated knowledge. For this purpose, a repertoire of about 140 relations and functions is provided, which detect typical semantic relations between concepts. Remote node attributes for aspects such as generalization, quantification and factuality are introduced which allow a precise description of the concept of meaning. Another feature can be encapsulated subnets in MultiNet and represented by a node of higher order, which itself may occur as an argument of relations and functions again.

The approach has been used in numerous practical applications of Natural Language Processing (NLP ). It was used for example for the implementation of natural language interfaces to the Internet, for question-answering systems through semantically annotated corpora with millions of records in deep -semantic search engines, for the semantic -based readability testing of texts as well as for the semantic -based detection of duplicates and plagiarism.

The MultiNet formalism is supported by various software tools. He also serves as a basis for the construction of a large semantic- based computer lexicon. The toolbox includes a parser for German, which transforms natural language expressions ( phrases, sentences or texts) in formal MultiNet expressions; a workbench for the knowledge engineer, the knowledge acquisition and automatic knowledge processing support; and a workbench for the computer lexicographer, the construction and management of large computer encyclopedias simplified.