Paper: SSMT:A Machine Translation Evaluation View To Paragraph-to-Sentence Semantic Similarity

ACL ID S14-2102
Title SSMT:A Machine Translation Evaluation View To Paragraph-to-Sentence Semantic Similarity
Venue Joint Conference on Lexical and Computational Semantics
Session
Year 2014
Authors

This paper presents the system SSMT measuring the semantic similarity between a paragraph and a sentence submitted to the SemEval 2014 task3: Cross-level Se- mantic Similarity. The special difficulty of this task is the length disparity between the two semantic comparison texts. We adapt several machine translation evalua- tion metrics for features to cope with this difficulty, then train a regression model for the semantic similarity prediction. This system is straightforward in intuition and easy in implementation. Our best run gets 0.808 in Pearson correlation. METEOR- derived features are the most effective ones in our experiment.