Software Development Effort Estimation Using Regression Fuzzy Models

Abstract

Software effort estimation plays a critical role in project management. Erroneous results may lead to overestimating or underestimating effort, which can have catastrophic consequences on project resources. Machine-learning techniques are increasingly popular in the field. Fuzzy logic models, in particular, are widely used to deal with imprecise and inaccurate data. The main goal of this research was to design and compare three different fuzzy logic models for predicting software estimation effort: Mamdani, Sugeno with constant output, and Sugeno with linear output. To assist in the design of the fuzzy logic models, we conducted regression analysis, an approach we call “regression fuzzy logic.” State-of-the-art and unbiased performance evaluation criteria such as standardized accuracy, effect size, and mean balanced relative error were used to evaluate the models, as well as statistical tests. Models were trained and tested using industrial projects from the International Software Benchmarking Standards Group (ISBSG) dataset. Results showed that data heteroscedasticity affected model performance. Fuzzy logic models were found to be very sensitive to outliers. We concluded that when regression analysis was used to design the model, the Sugeno fuzzy inference system with linear output outperformed the other models.

A New Approach to Software Effort Estimation Using Different Artificial Neural Network Architectures and Taguchi Orthogonal Arrays

“…These numerical values of 15 cost drivers in COCOMO81 and seventeen cost drivers in COCOMO2000 are increased to induce the trouble adjustment factor, that is, EAF. The performance of estimation [40] strategies is sometimes evaluated by many quantitative relation measurements of accuracy metrics [41] as well as RE (relative error), MRE (magnitude of relative error), MAE (mean absolute error) [42], and MMRE (mean magnitude of relative error) that are calculated as follows 12…”

Proposed Hybrid Approach to Predict Software Fault Detection

“…[14] surveyed on the resampling techniques, threshold, and ensemble algorithms on Naï ve Bayes, Random Forest and AdaBoost and he concluded that AdaBoost is effective classifier in achieving highest classification accuracy. Similarly, [15] used software metrics for predicting software module reuse and concluded by classifying reuse modules into four classes for fault prediction. [5] attempted to classify data using Naï ve Bayes classifier for predicting defect at an early stage using Bayes Theorem and achieved significant results in fault detection at inductive learning.…”

“…[16] used design metrics for object-oriented software and used Naï ve Bayes classifier for detecting faults at starting design phase as software development advances detecting faults is of high maintenance cost and time. (Nassif et al, 2019) used software metrics for predicting software module reuse and concluded by classifying reuse modules into four classes for fault prediction.…”

Proposed Intelligent Software System for Early Fault Detection

“…It has been observed that the criteria weights have large values when the considerations consist of larger differences in it, thus, reducing the entropy. The information entropy calculated using Renyi Entropy [23] is:…”