An approximation to the effect on learning a foreign language due to through Big Data collection from online language exams

Abstract

Artificial intelligence oriented to education (AIEd) allows the adequacy and / or adaption to the user’s learning itineraries through inductive processes based on the extraction of data obtained from the formative evidences that it generates throughout its school life. Big data, or massive data, is the storage of large amounts of data that can be analyzed by various procedures and allows us to find repetitive patterns or predictive formulas that can generate learning about ourselves and especially the network. In the case of the massive data that are generated through the use of tests in the learning and certification of knowledge of languages as a foreign language at the national level, we find that it might be useful to apply Big Data's processing methodologies in order to know better if the information Generated through the tests can improve or create new learning strategies or establish formal criteria in the design of the tests, theories of second language acquisition, or even educational policies. The novelty of the article focuses on establishing viable guidelines to apply the more generic concepts of Big Data in the specific context of the tests of language evaluation as a second language and where there is a priori a large amount of information to be processed at the educational level. The article shows some guidelines that could be applied in the mechanisms used in the extraction of educational data from large-scale language learning in the specific environment of language assessment tests as a foreign language.