Abstract The article focuses on the derivational perspective of metaphor studies. Derivation is regarded as a complex cognitive process, represented within speech activities. In this sense, derivation is viewed as a universal process of language units’ production according to the rules of text-formation. The basic feature of the derivational approach to the mechanism of metaphor is determined by the inner syntax, especially by the principle of contamination of two sentences – introductive and basic, which fulfill different functions. In this paper we shall present a theoretical account of metaphorisation as a universal derivational process controlled by means of such laws, as incorporation, contamination and compression. We take as basic the premise that metaphor is a more complicated process than it is described in traditional theories, since it is dependent on cognition and knowledge communication. In contrast to the traditional approaches, metaphor is regarded here as the result of combination of two pictures of the reality, referential and imaginative. We believe that derivatology generates a new knowledge about metaphor mechanism and metaphor modeling. Comparing to linear models of metaphor, the derivational model is considered to be a network model. The latest derivatological ideas about metaphor enrich the concept of metaphor taking into consideration that it has to be studied not in isolation, but within a broad frame of text, discourse, cognition and communication.
Are Neural Network Models Truly Effective at Forecasting? An Evaluation of Forecast Performance of Traditional Models with Neural Network Model for the Macroeconomic Data of G-7 Countries
