A probabilistic model for predicting software development effort

Authors:
Parag C. Pendharkar;Girish H. Subramanian;James A. Rodger
Affiliations:
Pennsylvania State University at Harrisburg;Pennsylvania State University at Harrisburg;Indiana University of Pennsylvania
Venue:
ICCSA'03 Proceedings of the 2003 international conference on Computational science and its applications: PartII
Year:
2003

Citing 8
Cited 0

Probabilistic reasoning in intelligent systems: networks of plausible inference

Probabilistic reasoning in intelligent systems: networks of plausible inference
A field study of scale economies in software maintenance

Management Science - Special issue: Frontier research on information systems and economics
Effort estimation and prediction of object-oriented systems

Journal of Systems and Software
Bayesian Analysis of Empirical Software Engineering Cost Models

IEEE Transactions on Software Engineering
A Vector-Based Approach to Software Size Measurement and Effort Estimation

IEEE Transactions on Software Engineering
Quantifying the Effects on Effort of Process Improvement

IEEE Software
Disaggregating and Calibrating the CASE Tool Variable in COCOMO II

IEEE Transactions on Software Engineering
An examination of some software development effort and productivity determinants in ICASE tool projects

Journal of Management Information Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

We use the naíve Bayes model to forecast software effort. A causal model is developed from the literature, and a procedure to learn Bayesian prior and conditional probabilities is provided. Using a data set of 40 real-life software projects we test our model. Our results indicate that the probabilistic forecasting models allow managers to estimate joint probability distribution over different software effort estimates. A software project manager may use the joint probability distribution to develop a cumulative probability distribution, which in turn may help the manager estimate the uncertainty that the project effort may be greater than the estimated effort.