The web, as we experience it nowadays, is heavily based on search engines such as Google or Yahoo!. These engines are the essential step to discover web-content that would, otherwise, only be available after too many clicks. The field of computer-science which serves as their theoretical basis is information retrieval. However, the main focus of information retrieval is on textual content, that is words and sentences. Little research has been done, however, both in terms of research or tools, for information retrieval regarding \it mathematical content on the web as can be seen, for example, in the overview of Robert Miner in the workshop Enhancing the Search of Mathematics. In this article, we present work done for the LeActiveMath learning environment which stores and presents semantically encoded mathematical content. We have adapted information retrieval techniques to this semantic content in order to offer to learners reasonably tolerant searchability for text, metadata, and formulae. Our efforts follow information retrieval principles stating the essential needs for fast response and easy query inputs.
