An Automatic Generation of Multiple-choice Cloze Questions Based on Statistical Learning

Since English expressions are different according to the genre, it is important for users to study with questions that are generated from sentences of the target genre. Although various questions are prepared, it is still not enough to satisfy various genres which users want to learn. On the other hand, when producing English questions, enough grammatical knowledge and vocabulary are needed, so it is difficult for inexperts to prepare English questions by themselves. In this paper, we propose an automatic generation system of multiple-choice cloze questions from English texts. Experiential knowledge is necessary to produce appropriate questions, so machine learning is introduced to acquire knowledge from existing questions. To generate the questions from texts automatically, the system (1) extracts appropriate sentences for questions from texts based on Preference Learning, (2) estimates a blank part based on Conditional Random Field, and (3) generates distracters based on statistical patterns of existing questions. Experimental results show our method is available for selecting appropriate sentences and appropriate for estimating the blank part.