CLUSTER BASED REGRESSION METHOD FOR SOFTWARE EFFORT ESTIMATION

Authors

  • Dr V.Vignaraj Ananth, Dr S.Srinivasan , M.Bhuvaneshwari

Abstract

Estimating the software development costs, budgets and resources such as the time and effort is
one of the most important activities in the software project management. The success of software
development depends very much on proper estimation of effort required to develop the software. Effective
software effort estimation techniques enable project managers to schedule software life cycle activities
properly. Software Effort Estimation is the process of predicting the most realistic amount of effort required
to develop or maintain software based on incomplete, uncertain and noisy input. The main research work
carried out in this paper is to accurately estimate the effort required in developing various software projects.
Before estimating the effort for the software, missing values in the datasets must be handled. In the proposed
method, the missing values problem in the dataset has been overcame by using k-means clustering
algorithm. The optimization of the effort parameters is achieved using the Linear Regression technique for
better prediction accuracy. Furthermore, performance of Linear Regression technique and Gaussian Process
technique are compared using well standard dataset with missing values. The experimental results show that
the Linear Regression with k-means clustering is performed better than the existing method in terms of
effort estimation accuracy.

Published

2020-06-30

Issue

Section

Articles