Date of Award
Dr. Jeff Matocha
Dr. Joey Dodson
Dr. Darin Buscher
This research presents novel algorithms which generate sentences in a natural language, using natural language generation techniques. The purpose of the algorithms is to benefit foreign language learning. As far as we can tell, ours is the first such research being done in the field. In creating the algorithms, we also developed a piece of software to showcase the work and allow testing by users. The main algorithm begins by generating sentence models by using one of two methods, namely modeled sentence generation and semantic sentence generation. Each of these have benefits and drawbacks, which the user must take into consideration when generating sentences. When the models are generated, they are filled in word by word using a conjugation algorithm. The completed sentences are then returned to the user and may then be translated. There is still much work to do before we will be satisfied with the algorithms, but our research shows that it is possible to use natural language generation techniques to benefit foreign language learning.
Kausler, Jacob, "Natural Language Processing for Foreign Language Learning" (2015). Honors Theses. 197.