Automatic Chinese Essay Scoring Using Connections between Concepts in Paragraphs

Abstract

Automatic essay scoring (AES) system is a very important research tool for educational studies. Many studies indicate that current AES systems should be able to analyze semantic characteristics of an essay and include more such features to score essays. This study proposes a novel method which uses the similarity between the paragraphic conceptual connections in different essays to predict the scores of essays. Preliminary experiments show that the paragraphic conceptual structure in an essay can be an efficient feature for scoring the essay.