Automated Chinese Essay Scoring using Vector Space Models

This paper presents experiments using several vector space models in Automated Essay Scoring (AES). Firstly, we compare four different Vector Space Models (VSM) which are the Word-based Vector Space Model (W-VSM), the Weight Adapted Word-based Vector Space Model (WAW-VSM), the Latent Semantic-based Vector Space Model (LS-VSM) and the Sequence Latent Semantic-based Vector Space Model (SLS-VSM). The results show that the WAW-VSM with the addition of word relation information is better than the W-VSM, while the SLS-VSM is also better than the LS-VSM by considering the sequence information in document representation. After that, we add some statistical surface features in the experiments. With the application of Support Vector Regression (SVR), the final machine score is generated. The correlation between the machine score and the human score reaches that between two human scores in average.