Conceptual Framework for Development of Computer Technology Supporting Cross-Linguistic Knowledge Discovery

The paper focuses on a fundamental problem, that of developing goal-oriented knowledge creation theory. Here we suggest a conceptual framework for development of a specific computer technology. This technology supports cross-linguistic knowledge discovery in a parallel corpus of Russian texts and their aligned French translations. We assumed that knowledge was evolved by a team of experts who thus filled a knowledge gap in contrastive grammar of Russian and French. The gap was identified through investigating the Russian-French parallel text corpus. The study is important because of a current need to reconsider linguistic education programs in the light of state-of-the-art corpus linguistic research. The knowledge gap was identified in the area of understanding equivalences between Russian and French grammatical constructions. To establish baseline reference conditions for this computer technology development, we observed how experts worked and how new concepts emerged. The concepts are thought to be knowledge parts that come from some denotata. The denotata are Russian-French parallel text fragments. Experts analyze them to identify and describe types of equivalences between Russian and French grammatical constructions. A typology of those constructions is thus developed. We establish baseline reference conditions as follows: Experts' interaction is goal-oriented, Experts apply an iterative identification procedure for defining equivalences between Russian and French grammatical constructions, Knowledge discovering processes are irreversible during the iterative identification procedure, Results of corpus-based knowledge discovery are the personal concepts that belong to a knowledge space - the mental medium, Experts identify and describe their own personal concepts as results of semantic interpretation of denotata, the denotata being Russian-French parallel text fragments, Experts jointly coordinate their personal concepts as parts of emerging knowledge and generate collective concepts interpreting Russian-French parallel text fragments, Russian-French parallel text fragments are new knowledge sources and belong to the text corpus as parts of the social information medium. The proposed conceptual framework is based on these conditions and includes descriptions of following items: Mental medium, social information medium, digital medium and forms of human-computer interfaces, Individual, group and conventional concepts as signifieds in the mental medium, along with their computer codes named as semantic codes of the digital medium, Knowledge sources as denotata, that are Russian-French parallel text fragments into the social information medium, and their computer codes named as object codes of the digital medium. To show feasibility of the proposed conceptual framework, our team designed a new computer technology supporting cross-linguistic knowledge discovery. The technology is based on three forms of human-computer interfaces and knowledge creation models. We give a few examples to show what new knowledge could be discovered with the aid of the technology.