Finding Hidden Relationships Between Medical Concepts by Leveraging Metamap and Text Mining Techniques

Text is one of the most common ways to store data in this computerized world. At a glance, it may seem that those data are not interconnected. But in reality, data can have hidden connections. Therefore, in this research, a new model has been presented that can find hidden relationships between two medical concepts by using MetaMap and appropriate text-mining techniques. Specifically, the model creates a new comprehensive index structure and can find cross-document hidden links connecting topics of interest that most existing approaches have ignored. Experiments show the effectiveness of the proposed model in discovering new connections between topics.