*Paris is Rain. or It is raining in Paris?: Detecting Overgeneralization of Be-verb in Learner English.

This paper addresses the detection of overgeneralization of be-verb found in learner English. It is an error where the subject and complement are not semantically equivalent in a be-verb sentence as in *Paris is rain. This type of error often appears in the writing of learners whose native language has a be-verb equivalent that has usages other than those which English be-verb does. This paper presents a method for detecting overgeneralization of be-verb by predicting through word embeddings whether a given subject and complement pair is semantically equivalent or not. It also presents an effective and efficient way of determining the hyperparameters in the method. Experiments show that the present method outperforms four baseline methods based on corpus statistics and WordNet ontology despite the fact that it is a rather simple method. Looking into the detection results reveals the performance limitations of the proposed method.