Dilated Cardiomyopathy (DCM) is one of the cardiovascular diseases (CVDs) that is the root cause leading towards other CVDs such as Arrhythmias and Myocardial infarction (MI). The current methods available for the detection of DCM are less accurate, inefficient and are not effective to be used for diagnostic purposes. This study focuses on the detection of DCM using Pulse Plethysmograph (PuPG) signals, which is an efficient, cheap and relatively new method as compared to the existing practices. A total of 44 subjects were involved in this work for data acquisition, out of which 22 were suffering from DCM and the rest were normal subjects, the PuPG signal data was acquired from the index finger of these subjects. Discrete Wavelet Transform (DWT) was employed for the denoising the PuPG signal. Extensive experimentation resulted in the selection of three features which were giving the best intraclass difference through K-nearest neighbor (KNN) and Support Vector Machines (SVM). An accuracy of 99.7%, specificity of 99% and sensitivity of 100% was achieved using Gaussian kernel for SVM.