Recognizing Severity of Disability Using Gradient Boosted Trees

The amount of data being stored daily is enlarging at an explicit velocity. Medicine is one of these domains where new knowledge is accumulated at a daily basis. Globally the healthcare sector is abundant with data. Hence using data mining techniques in this area seems indispensable. Data mining brings a set of tools and techniques that can be applied to discover hidden patterns that provide healthcare professionals an additional source of knowledge for making decisions. This paper intends to use data mining techniques to gain knowledge from a dataset, which is provided by Welfare Organization in Tabriz, Iran. The purpose of this paper is to predict the severity of disability of each patient, using the symptoms and cause of disability. For labeling the instances, a group of doctors has determined each patient’s severity of disability. A number of experiments have been performed to compare the performance of different data mining techniques and the results illustrate that Gradient Boosted Trees works better among other techniques such as Decision Tree, Random Forest, Naive Bayes, K-Nearest Neighbors and etc.

Keywords:

Data Mining , Classification , Severity of Disability , Gradient Boosted Trees