The existence of “missing links” of semantic information in natural language bases stored in digital format, de?nes the Semantic Gap in the context of Natural Language Processing. Semantic Networks provide an easy way for visualizing aknowledge base and also e?cient algorithms for inferring properties of an object based on its category membership and have been popular for representation of structured knowledge.In this work we evaluate how lexical/semantic resources, with a focus on semantic networks, can be used and combined to contribute on bridging this semantic gap between concepts and also evaluate the how these resources can contribute to improve the results of sentence alignment for textual inference.