Abstract: A decision-tree is a structure which can represent any algorithm in a tree or even graph with the nodes and branches with some associated outcome in terms of weight and probability. Basically here, we are using large datasets gathered in clinical records related to a patient’s health care issues and their medical reports to predict his or her future health diagnosis and accordingly recommend the required medication. The recommender tool being implemented in Hadoop framework uses popular classification algorithm named as C4.5. We are trying to improve the performance of the decision-tree algorithm utilizing the appropriate Map-reduce model and forward it to Hadoop framework in addition to bagging method additionally with random-subspaces in ensemble classification in order to improve efficiency and scalability.

Keywords: C4.5, algorithm, decision-tree, ensemble classification, Hadoop framework, Map-reduce, Recommender tool, PHR, HRS.