A Cardiotocographic Classification using Feature Selection: A comparative Study

Cardiotocography is a series of inspections to determine the health fetus in pregnancy. The inspection process carried out by recording baby's heart rate information whether healthy condition or contrarily. In addition, uterine contractions are also used fetus. Fetal classified into 3 conditions namely normal, suspect, and pathological. This paper was performed compare classification algorithm for diagnosing result cardiotocographic inspection. An experimental scheme using feature selection not it. CFS Subset Evaluation, Info Gain, Chi-Square select best which correlated each other. data set obtained from UCI Machine Learning repository available freely. To find performance algorithm, this study uses an evaluation matrix precision, Recall, F-Measure, MCC, ROC, PRC, Accuracy. results showed that all algorithms can provide fairly good classification. However, combination Random Forest Gain Feature Selection gives with accuracy 93.74%.