This paper proposes a supervised learning method for detecting a semantic relationbetween a given pair of named entities, which may be located in different sentences.he method employs newly introduced contextual features based on Salient Referent List as well as conventional syntactic and word-based features.These features are organized as a tree structure and are fed into a boosting-based classification algorithm.Experimental results show the proposed method outperformed prior methods, andincreased precision and recall by 11.3% and 14.2%.