Application of big data analysis with decision tree for the foot disorder

In medical field, massive data sets is generated by rapid development of hospital information system. For analysis of thesemedical big data, this study showed analysis process of the clinical data to acquire significant information effectively between the foot disorder groups and biomechanical parameters related to symptom by developing a prediction model of the decision tree. The first clinical health records of 1523 patients diagnosed with foot disorder were used for analysis, in total 6610 records. The dependent variable in the analysis data was consisted of five complex disorder groups, and the independent variable was composed of 24 attributes. The decision tree was applied to analyze pattern of the foot disorder. The measured prediction rate was Correct: 72.96% and Wrong: 27.04% in the training data, and Correct: 68.66% and Wrong: 31.34% in the test data. As a result of analysis on the five foot complex foot disorder groups by using C5.0 algorithm, 12 rules were generated. To improve accuracy of classification, the detailed preprocessing and other data mining algorithms will be applied from now on.