Detection and analysis of software defects at a very early stage is very much essential in the domain of software engineering. It also influences the decision-making process related to allocation of resources for evaluation or verification. Software quality assurance can be defined as a significant phenomenon for the implementation of various machine learning techniques in defect detection. These techniques basically emphasize on single product-based software defects rather than the multi-product-based defects. Software reliability prediction models are used to predict the fault rate of the software systems using machine learning models. A large number of traditional reliability measures are used to test the software faults in the debugging and testing process. Most of the traditional machine learning based fault prediction models are integrated with standard software reliability growth measures for reliability severity classification. However, these models are used to predict the reliability level of binary class with less standard error. In this paper, a hybrid support vector regression-based non-linear parametric growth measure is implemented on the training fault datasets. Experimental results are simulated on various reliability datasets with different configuration parameters for fault prediction