Kohanduv veebirakendus eesti keele sõnavara ja grammatika õppimiseks

Abstract

The goal of this Bachelor’s thesis was to create a web-based study game for learning Estonian cases and synonyms. The application adapts to the answers given by the participant. It is suitable for pupils from elementary to high school and, in addition, is appropriate for people studying Estonian as a non-native language. For the creation of synonym tests the following natural language tools were used: two corpuses, a morphological analyser, Estonian Wordnet, a list of unsuitable words and a Word2Vec machine learning model. To achieve adaptation to the user in the game, a simplification of the Leitner system was built. A description of the application, test generation algorithm and adaptation algorithm are included in the thesis.