Get our free extension to see links to code for papers anywhere online!


Applying Part-of-Seech Enhanced LSA to Automatic Essay Grading

Add code

Oct 19, 2006
Tuomo Kakkonen, Niko Myller, Erkki Sutinen


Share this with someone who'll enjoy it:


Latent Semantic Analysis (LSA) is a widely used Information Retrieval method based on "bag-of-words" assumption. However, according to general conception, syntax plays a role in representing meaning of sentences. Thus, enhancing LSA with part-of-speech (POS) information to capture the context of word occurrences appears to be theoretically feasible extension. The approach is tested empirically on a automatic essay grading system using LSA for document similarity comparisons. A comparison on several POS-enhanced LSA models is reported. Our findings show that the addition of contextual information in the form of POS tags can raise the accuracy of the LSA-based scoring models up to 10.77 per cent.

* Proceedings of the 4th IEEE International Conference on Information Technology: Research and Education (ITRE 2006). Tel Aviv, Israel, 2006 


   Access Paper Source



Share this with someone who'll enjoy it: