Building Quality Estimation models with Fuzzy Threshold Values
نویسندگان
چکیده
This work presents an approach to circumvent one of the major problems with techniques to build and apply software quality estimation models, namely the use of precise metric thresholds values. We used a fuzzy logic based approach to investigate the stability of a reusable class library interface, using structural metrics as stability indicators. To evaluate this new approach, we conducted a study on three versions of a commercial C++ class library. The obtained results are very promising when compared to those of two classical machine learning (ML) approaches, Top Down Induction of Decision Trees and Bayesian classifiers. RÉSUMÉ: Ce travail présente une approche pour résoudre un des principaux problèmes reliés aux techniques de construction et d’application des modèles d'estimation de la qualité de logiciel, à savoir l'utilisation des valeursseuils précises pour les métriques. Nous avons utilisé une approche basée sur la logique floue pour étudier la stabilité de l’interface des bibliothèques de classes. Nous nous sommes basés sur les métriques structurelles comme indicateurs de stabilité. Pour évaluer cette nouvelle approche, nous avons entrepris une étude sur trois versions d'une bibliothèque de classes commerciale écrite en C++. Les résultats obtenus sont très prometteurs comparativement à ceux de deux approches classiques d’apprentissage,TDIDT et classificateurs bayésiens.
منابع مشابه
Extending Software Quality Predictive Models Using Domain Knowledge
Current methods to build software quality estimation models suffer from two problems: The use of precise threshold values and their interpretation in the absence of formal models, and the crudeness of the derived rules which can only serve to build naïve models. In this position paper, we describe a novel approach to alleviate both problems. We propose to build fuzzy decision processes that com...
متن کاملApplication of Soft Computing Methods for the Estimation of Roadheader Performance from Schmidt Hammer Rebound Values
Estimation of roadheader performance is one of the main topics in determining the economics of underground excavation projects. The poor performance estimation of roadheader scan leads to costly contractual claims. In this paper, the application of soft computing methods for data analysis called adaptive neuro-fuzzy inference system- subtractive clustering method (ANFIS-SCM) and artificial neu...
متن کاملComparison of Linear and Threshold Models for Estimation Genetic and Phenotypic Parameters of Success of Conception at First Service and Inseminations to Conception in Holstein Cattles in East Azarbayjan Province
In this research genetic and phenotypic parameters were estimated using linear and threshold models, for reproductive traits, data from 6 large industrial dairy herd of East Azerbaijan province collected by Agriculture Jihad Organization during 10 years (2001-2010). Best linear unbiased predictions of traits breeding values were estimated using Restricted Maximum Likelihood method by WOMBAT sof...
متن کاملComparison of Linear and Threshold Models for Estimation Genetic and Phenotypic Parameters of Success of Conception at First Service and Inseminations to Conception in Holstein Cattles in East Azarbayjan Province
In this research genetic and phenotypic parameters were estimated using linear and threshold models, for reproductive traits, data from 6 large industrial dairy herd of East Azerbaijan province collected by Agriculture Jihad Organization during 10 years (2001-2010). Best linear unbiased predictions of traits breeding values were estimated using Restricted Maximum Likelihood method by WOMBAT sof...
متن کاملAN INTELLIGENT INFORMATION SYSTEM FOR FUZZY ADDITIVE MODELLING (HYDROLOGICAL RISK APPLICATION)
In this paper we propose and construct Fuzzy Algebraic Additive Model, for the estimation of risk in various fields of human activities or nature’s behavior. Though the proposed model is useful in a wide range of scientific fields, it was designed for to torrential risk evaluation in the area of river Evros. Clearly the model’s performance improves when the number of parameters and the actual d...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2001