Investigating Effort Prediction of Software Projects on the ISBSG Dataset

Abstract

Many cost estimation models have been proposed over the last three decades. In this study, we investigate fuzzy ID3 decision tree as a method for software effort estimation. Fuzzy ID software effort estimation model is designed by incorporating the principles of ID3 decision tree and the concepts of the fuzzy settheoretic; permitting the model to handle uncertain and imprecise data when presenting the software projects.MMRE (Mean Magnitude of Relative Error) and Pred(l) (Prediction at level l) are used, as measures of prediction accuracy, for this study. A series of experiments is reported using ISBSG software projects dataset. Fuzzy trees are grown using different fuzziness control thresholds. Results showed that optimizing the fuzzy ID3 parameters can improve greatly the accuracy of the generated software cost estimate.

Software Development Effort Estimation Techniques: A Survey

“…A. -Pred (25) The NBC model was used to be as useful as the SWR model in terms of prediction accuracy using Pred (25) [20]. M. Kumar, et al [18] -Linear Regression, -Random Forest, -Multi-layer perceptron.…”

Systematic Review Study of Decision Trees based Software Development Effort Estimation

“…[42] Can deal with imprecision and uncertainty. [32], [33], [34] The ability to learn from past finished projects and have predictive power.…”

Cluster-based fuzzy regression trees for software cost prediction

“…Hence, the researchers explain the uncertainty's sources or propose solutions to manage imprecision in software effort prediction environments. Mainly, in [5], [15], [16], the authors investigated the use of the fuzzy ID3 decision tree in SDCP by incorporating the principle of fuzzy theory. Idri and Elyassami [15], the experiments show that based on Pred (25) and MMRE, the Fuzzy ID3 outperforms crisp ID3, CART, and C4.5 over the TUKUTUKU dataset.…”

“…Idri and Elyassami [15], the experiments show that based on Pred (25) and MMRE, the Fuzzy ID3 outperforms crisp ID3, CART, and C4.5 over the TUKUTUKU dataset. Moreover, in Elyassami [16], the use of the ISBSG dataset enhances the performance. Also, in Elyassami and Idri [17] the authors are interested in testing the accuracy of fuzzy C5 and the effect of the confidence factor of pruning on the accuracy of fuzzy C5 using the ISBSG dataset [18].…”

“…Nevertheless, the proposed models are still suffering from some weaknesses. For example, in [16], the author does not compare with other SDCP methods, which hinders the validity of the results. Also, in [5], the authors compare the proposed models just with the crisp version of the models or with just one model like in [17], [18], which is not enough for generalizing the results and affirming their superiority.…”