ARPN Journal of Science and Technology::A Novel Imputation Technique for Software Effort Estimation

Statistical analysis is greatly hindered with missing information. It represents a loss of key data, but worse, it can introduce biased results in the analysis. A way to rectify the problem of missing data is to employ a sound method of imputation, a way to replace missing values with reasonable estimates. Their exists variety of estimation models like SLIM, COCOMO and other models like machine learning models which require accurate inputs for estimation of effort. It is very difficult to obtain such accurate information at the early development stage of a software project. These issues led to the introduction of variety of techniques to solve the relevant issues. The proposed approach is a hybrid approach which works based on Genetic algorithm and empirical estimation model COCOMO II.