Control charts are widely used in manufacturing to monitor whether a process is in control to ensure the consistency of the process. However, on emergence of any unexpected occurrence in a process, a control chart is unable to recognize the pattern of the dataset. Over the years, different methods have been developed to recognize unnatural control chart patterns. A methodology for recognizing control chart patterns is statistical correlation measure that measures correlation between a test dataset and a reference dataset. Support vector machine (SVM) has been the most popular model in recent years to recognize control chart patterns. However, computational effort in SVM model is high and training process of the model is complex and expensive. In this paper, these two methodologies have been applied to different datasets to recognize unnatural patterns. Their performance have been measured and compared. The comparison shows that SVM model performs substantially better than statistical correlation method in terms of accuracy in the output. Statistical correlation model is less expensive and associated with a simpler algorithm than the SVM model, but not as accurate as SVM model. On the other hand, the computational time required for SVM model is higher than the statistical correlation model.