Under Skewed Class Distribution Data (SCDD) Real-world applications often involve highly skewed data in decision outcomes. Such a highly skewed class distribution problem, if not properly addressed, would imperil the resulting learning effectiveness. We use one hospital Central Venous Catheter Insertion data for the training data. Among them find CVP-tip attribute are skewed class distribution. The paper uses under-sampling over-sampling and cost-sensitivity to solve SCDD problem. And then use C4.5 algorithms to build up classification model for the central venous catheter insertion. Verify via the expert that use this model can effectively help the medical expert to extract out it about Central Venous Catheter Insertion good knowledge rules.