In this paper we propose and carefully evaluate the application of an information theoretic approach for the detection of hypernyms for financial concepts. Our algorithm is based on the application of sparse word embeddings, meaning that – unlike in the case of traditional word embeddings – most of the coefficients in the embeddings are exactly zero. We apply an approach that quantify the extent to which the individual dimensions for such word representations convey the property that some word is the hyponym of a certain top-level concept according to an external ontology. Our experimental results demonstrate that substantial improvements can be gained by our approach compared to the direct utilization of the traditional dense word embeddings. Our team ranked second and fourth according to average rank score and mean accuracy that were the two evaluation criteria applied at the shared task.