DISCO allows to retrieve the semantic similarity between arbitrary input words, and lists the semantically most similar words for an input word. It consists of a multilingual database with a Java API. Pre-computed word spaces are available for a number of languages, so DISCO can be used off the shelf.
TLA-team: Module for semantic comparison of words. Can be used as a command line tool or as a library in java applications (under the Apache license), with a fairly well documented API. Corpora available for a number of major languages. A plugin for Protégé is available as well.