Aim : The aim of the proposed system is to predict the software effort estimation accurately using a support vector machine and comparing accuracy with k nearest neighbour. Materials and Methods:The discussion is about the estimation of software effort by using machine learning algorithms such as support vector machine and k nearest neighbour. Here pretest power analysis was carried out at 80% and the sample size and the iteration values are 20 and 15 with the significant value being 0.002 Results: The software effort estimatedusing support vector machine and k nearest neighbour with the accuracy of 86.37% and 72.23% respectively. Conclusion:support vector machine performs significantly better than k nearest neighbour

Support Vector Machine, K-Nearest Neighbour, Novel Biomarker, software Effort, Artificial Intelligence, Environmental Engineering