Graph Mining Healthcare Approach: Analysis and Recommendation

Abstract

Machine learning and computational intelligence have facilitated the development of recommendation systems for a broad range of domains. Such recommendations are based on contextual information that is explicitly provided or pervasively collected. Recommendation systems often improve decision-making or increase the efficacy of a task. An obvious application is a person’s physical health where it is advantageous to increase the number of healthy cells in the body and destroy cancerous cells (wherein cancer is your opponent), we can learn how to predict positive outcomes for such scenarios. Herein we show how frequent and discriminative subgraph mining can be employed to analyze a collection of healthcare dataset cases and make recommendations about sequences of actions that should take, as well as should not take, be made to increase the chances of a patient's recovery in the near future. As proof of concept, we present the results of an experiment that utilizes our strategy for one particular healthcare dataset, MIMIC.