We present an empirical study on the use of semantic information for Concept Segmentation and Labeling (CSL), which is an important step for semantic parsing. We represent the alternativeanalyses output by a state-of-the-art CSL parser with tree structures, which we rerank with classifier trained on two types of semantic tree kernels: one processing structures built with words, concepts and Brown clusters, and another one using semantic similarity among the words composing the structure. The results on a corpus from the restaurant domain show that our semantic kernels exploiting similarity measures outperform state-of-the-art rerankers.