This paper presents an application of some Inductive logic programming (ILP) techniques for checking user’s answer correctness in a Computer-Aided Language Learning (CALL) system STyLE (Scientific Terminology Learning Environment). STyLE supports adaptive learning of English terminology with a target user group of non-native English speakers. In STyLE are implemented many original features that make this system intelligent and adaptive, but we will focus only on one of them: supporting learner-system communication in Natural Language (NL). The proposed ILP system Rich is used for generation of least generalization(LG) and greatest specialization(GS) of the set of possible correct answers of a given question to the user from the system. The user’s answer is correct if it is between LG and GS of the correct answers’ set.