Research on Recognition Algorithm of English Gerund Collocation Errors Based on Data Classification

The most common verb matching errors that English learners are prone to are found in using gerunds. After analyzing the errors in CLEC, we provide a standard model to correct the errors for Chinese language learners. Firstly, a standard verb language matching library is constructed, and then a way to measure the similarity between target matches is provided. Then, a rough set of similar matches is obtained based on the statistical similarity between target matches and the standard matching library. Finally, a classification method is used to filter out matches that cannot be classified as the correct type in the similarity matching set, resulting in a candidate result set. Finally, the candidate result set is ranked based on the language module, and the final correction opinion is obtained. On the standard detection set constructed using BNC, this method of combining similarity judgment and contextual features has a significant effect on finding and correcting verb-matching errors.