While ontologies are widely accepted internationally as knowledge management mechanism across disciplines, the ability to reason over knowledge bases regardless of the natural languages used in them has become a pressing issue in digital content management. To enable knowledge sharing and reuse, ontology mapping techniques must be able to work with otherwise isolated ontologies that are labelled in diverse natural languages. Machine translation techniques are often employed by cross-lingual ontology mapping approaches to turn a cross-lingual mapping problem into a monolingual mapping problem which can then be solved by state of the art monolingual ontology matching tools. However in the process of doing so, complications introduced by machine translation tools can compromise the performance of the subsequent monolingual matching techniques. In this paper, a novel approach to improve the quality of cross-lingual ontology mapping is presented and evaluated. The proposed approach adopts the pseudo feedback technique that is similar to the well understood relevance feedback mechanism used in the field of information retrieval. It is shown through the evaluation that the pseudo feedback feature can enhance the effectiveness of machine translation and monolingual matching techniques in a cross-lingual ontology mapping scenario.