Semantic interpretation of compound nominalization using automatic extracted paraphrase patterns

The interpretation of nominal compounds is one of the most difficult problems in natural language processing. This paper proposes a new algorithm for the automatic interpretation of compound nominalizations in Chinese. A maximum entropy model is applied to label four coarse-grained roles involved in compound nommializations. The classification features used are automatic extracted functional lexicalized paraphrase patterns. The experimental result shows that the method is superior than previous works that use manually coded paraphrase patterns.