This research investigated the ‘applicability of machine learning algorithms based on decision trees to the information retrieval process’. A number of experiments were conducted on records generated from a collection. Based on a random selection of records, decision trees were created for each record. The decision tree could then categorize the record as relevant or irrelevant bassed on the query used to retrieve the document. In order to evaluate the decision tree each record was presented to the tree, which then produced an ‘irrelevant’ response to the original query. This was then compared with the performance of an inverted indexing system to evaluate levels of performance.