Software Quality Prediction Research Articles

A systematic review of hyperparameter tuning techniques for software quality prediction models

BACKGROUND: Software quality prediction models play a crucial role in identifying vulnerable software components during early stages of development, and thereby optimizing the resource allocation and enhancing the overall software quality. While various classification algorithms have been employed for developing these prediction models, most studies have relied on default hyperparameter settings, leading to significant variability in model performance. Tuning the hyperparameters of classification algorithms can enhance the predictive capability of quality models by identifying optimal settings for improved accuracy and effectiveness. METHOD: This systematic review examines studies that have utilized hyperparameter tuning techniques to develop prediction models in software quality domain. The review focused on diverse areas such as defect prediction, maintenance estimation, change impact prediction, reliability prediction, and effort estimation, as these domains demonstrate the wide applicability of common learning algorithms. RESULTS: This review identified 31 primary studies on hyperparameter tuning for software quality prediction models. The results demonstrate that tuning the parameters of classification algorithms enhances the performance of prediction models. Additionally, the study found that certain classification algorithms exhibit high sensitivity to their parameter settings, achieving optimal performance when tuned appropriately. Conversely, certain classification algorithms exhibit low sensitivity to their parameter settings, making tuning unnecessary in such instances. CONCLUSION: Based on the findings of this review, the study conclude that the predictive capability of software quality prediction models can be significantly improved by tuning their hyperparameters. To facilitate effective hyperparameter tuning, we provide practical guidelines derived from the insights obtained through this study.

Information technology for prediction of software quality level

Today, there is a contradiction between the rapid increase in the complexity and size of modern software while increasing responsibility for the performance of their functions, the increasing requirements of customers and users to the quality and efficiency of software use and the imperfection of models, methods, tools of predicting software quality at the early stages of the life cycle.Therefore, the task of predicting the software quality level based on requirements is relevant. The aim of this study is to solve this task by developing information technology for prediction software quality levels based on requirements. The proposed information technology for prediction software quality level based on requirements provides analysis of quality attributes in requirements, reflects the dependence (equations) of quality characteristics on attributes, forms a quantitative assessment of quality characteristics, reflects the dependence (equation) of quality on its characteristics, forms a quantitative assessment of quality, performs quality level prediction, provides all the listed services simultaneously and the model, methods, and tools underlying the IT belong to common methodological approaches and are integrated. The developed system of predicting the software quality level based on requirements provides the user with predicted estimates of eight software quality characteristics, geometric interpretation of the software quality characteristics’ values, a comprehensive indicator of the predicted software quality, and a conclusion about the future software quality level. On the basis of this, it is possible to compare sets of requirements for software and make a reasonable choice of a set of requirements for further implementation. The information technology and the system of predicting the software quality level based on requirements, developed in this paper, provide the possibility of comparing sets of requirements for software, justified selection of requirements for further implementation of quality software (as experiments have shown, this is only one of the four proposed sets), and rejection or revision of unsuccessful sets of requirements that cannot be used to develop quality software.

Experimental Investigation of Machine Learning Techniques for Predicting Software Quality

Experimental Investigation of Machine Learning Techniques for Predicting Software Quality

Prediction of software quality with Machine Learning-Based ensemble methods

Prediction of software quality with Machine Learning-Based ensemble methods

SOFTWARE QUALITY PREDICTION USING MACHINE LEARNING TECHNIQUES AND SOURCE CODE METRICS: A REVIEW

Software quality prediction is the Machine Learning (ML) based technique in which ML models are trained using historical data. Output from these quality models can be used by software experts in the early phase of software development for improving the quality of software by controlling the various quality attributes like maintainability, reliability, security issues of software etc. In this study a systematic review of studies from 2005 to 2021 is performed. Studies that use ML techniques and source code metrics for Software Quality Prediction (SQP) are included for review. Study assesses the commonly used machine learning techniques and source code metric for SQP. Commonly used datasets, feature selection techniques and commonly used performance measures in software quality prediction are also assessed. In this paper 53 primary studies are selected for systematic review. Results of this study prove that Bayesian Learning (BL), Regression, Ensemble Learning (EL), Decision Tree (DT) and Support Vector Machine (SVM) are most commonly ML techniques used for quality prediction which comprises 58%, 52%, 41%, 32%, and 32% of the overall studies respectively. It is also assessed that NASA, PROMISE, Apache, Mozilla Firefox and Eclipse are the most commonly used datasets for training and testing the SQP models. LOC, CC, CBO, RFC, WMC, LCOM, DIT and NOC are among the most commonly used source code metrics in SQP. Based on the results from the selected studies it is concluded that ML techniques and source code metrics have the ability to improve the overall quality of the software.

A Software Quality Prediction Model for Aspect Oriented System using Neuro-Fuzzy Approach

Nowadays, the usage of software has increased exponentially in various fields like education systems, industries, health systems, and many others. Various software architectures are already available in the market e.g. modular oriented, component-based, object-oriented, aspect-oriented, etc. Aspect-Oriented (AO) system software has gained much attention due to its superior features to the aforementioned systems. However, AO systems face the challenges of being complex and hard testing environment, a quality assessment of these systems is necessary. In this paper, a software quality estimation model for aspect oriented system using neuro-fuzzy approach has been developed. For which, a detailed study on aspect oriented systems has been accomplished in terms of various attributes affecting the quality of these software. In this paper, a framework of software quality prediction model has been designed using the adaptive neuro-fuzzy inference engine (ANFIS) approach. Data of 200 software pieces have been collected in this study where 150 software data is used to train the ANFIS model whereas 50 software data is used for testing purposes. The quality estimated by the proposed ANFIS model is compared with the actual quality of these software data and quantitative analysis is performed in terms of error measures. Finally, it was found that the proposed ANFIS model worked better in terms of MSE, MRE, MARE, MBRE, and MIBRE error measures.

NEURAL-NETWORK MODEL OF SOFTWARE QUALITY PREDICTION BASED ON QUALITY ATTRIBUTES

The paper proposes a neural-network model of software quality prediction based on quality attributes. The proposedmodel differs from the known models, because it provides considering the importance of each quality attribute and their interactionwithin each software quality characteristic. The artificial neural network (ANN) outputs correspond to the values of software qualitycharacteristics (functional suitability, performance efficiency, usability, reliability, compatibility, security, maintainabi lity, portability).The artificial neural network (ANN) outputs make it possible assessing the total impact of quality attributes on software qualitycharacteristics

Software Defect Prediction and Software Quality Assessment Using Dlr-Lvq and Fuzzy Rules

Recently, Software development has been considerably grown. Fault in the software causes fault and interrupts the output. Characteristics like these make it much challenging to avert software flaws. Spontaneously forecasting the amount of flaws within the software modules is essential and also can assist developers to proficiently allot restricted resources. Recently, numerous Software Defect Prediction (SDP) techniques are developed. But, the accuracy and time consuming challenges still remain to be solved. Also, a few top-notch techniques don't properly classify the software whereas it is a needed metric to ensure quality standards. This work proffers a novel Decaying Learning Rate – Learning vector Quantization (DLR-LVQ) classifier to forecast the software defect. The proposed methods consist of the following steps: redundant data removal, feature extraction (FE), feature oversampling, data normalization, defect prediction (DP), and quality prediction. The proposed DLR-LVQ’s attained outcome is assessed with the existent methodologies. The outcomes exhibit that the methodology proposed attains efficient classification outcomes are examined. Keywords: Software Defect Prediction (SDP), Non defective software quality prediction, BM-SMOTE, Decaying Learning Rate, Learning Vector Quantization, Fuzzy rules, HDFS and Map Reduce.

Software quality prediction using machine learning

BackgroundThe demand for software is increasing every day in various fields. Software developers put more effort to develop and test the quality of the software and verify its reliability before it is released. High-quality software modules were developed to allow others to reuse the components. Purpose: This paper provides information to researchers in the software quality prediction field based on machine learning algorithms. Methodology: Most of the machine learning techniques and the relevant software metrics used in many high-quality papers published between 2010 and the end of 2021 have been analysed. Findings: Machine learning techniques are the most suitable for predicting software quality. Most of the researchers' interest was to predict the reliability of software. The detailed analysis enables researchers to choose the best way to plan their research and to make a good contribution in the field of software quality prediction.

Qualimetry Essentials Applied to Embedded Software Development

ABSTRACTWith the initial goal of filling gaps in the definition, assessment, control, and prediction of automotive embedded software quality, including standards and regulatory compliance constraints, a three‐year research study was conducted. The multiple contributions of the study to theoretical and applied qualimetry have helped to define a theoretical framework for quality control and monitoring of engineering and product development processes. Then, demonstrating the practicality of this solution, this study was applied on a real example from the automotive industry. This framework initially developed in the scope of embedded software is also applicable to systems and embedded systems, and makes these contributions have an impact on all industrial sectors.

Prediction of software vulnerability based deep symbiotic genetic algorithms: Phenotyping of dominant-features

The detection of software vulnerabilities is considered a vital problem in the software security area for a long time. Nowadays, it is challenging to manage software security due to its increased complexity and diversity. So, vulnerability detection applications play a significant part in software development and maintenance. The ability of the forecasting techniques in vulnerability detection is still weak. Thus, one of the efficient defining features methods that have been used to determine the software vulnerabilities is the metaheuristic optimization methods. This paper proposes a novel software vulnerability prediction model based on using a deep learning method and SYMbiotic Genetic algorithm. We are first to apply Diploid Genetic algorithms with deep learning networks on software vulnerability prediction to the best of our knowledge. In this proposed method, a deep SYMbiotic-based genetic algorithm model (DNN-SYMbiotic GAs) is used by learning the phenotyping of dominant-features for software vulnerability prediction problems. The proposed method aimed at increasing the detection abilities of vulnerability patterns with vulnerable components in the software. Comprehensive experiments are conducted on several benchmark datasets; these datasets are taken from Drupal, Moodle, and PHPMyAdmin projects. The obtained results revealed that the proposed method (DNN-SYMbiotic GAs) enhanced vulnerability prediction, which reflects improving software quality prediction.

Application of metaheuristic techniques in software quality prediction: a systematic mapping study

Application of metaheuristic techniques in software quality prediction: a systematic mapping study

Application of metaheuristic techniques in software quality prediction: a systematic mapping study

Application of metaheuristic techniques in software quality prediction: a systematic mapping study

An Empirical Study on Software Fault Prediction Using Product and Process Metrics

Product and process metrics are measured from the development and evolution of software. Metrics are indicators of software fault-proneness and advanced models using machine learning can be provided to the development team to select modules for further inspection. Most fault-proneness classifiers were built from product metrics. However, the inclusion of process metrics adds evolution as a factor to software quality. In this work, the authors propose a process metric measured from the evolution of software to predict fault-proneness in software models. The process metrics measures change-proneness of modules (classes and interfaces). Classifiers are trained and tested for five large open-source systems. Classifiers were built using product metrics alone and using a combination of product and the proposed process metric. The classifiers evaluation shows improvements whenever the process metrics were used. Evolution metrics are correlated with quality of software and helps in improving software quality prediction for future releases.

Hybrid Fuzzy Knowledge Based Prediction Model for the Software Development and Maintenance Quality in Software Engineering Approach

The main arena of Software Engineering development with ood design, development, coding, testing, implementation, deployment of the software, finally maintaining the software with good functionality. For the development of software many organizations are investing more and more budget in their revenue. Software Engineering development has several categories of data presented in software engineering such as Graphical User Interface, Usage graphs, writing text, realities and images. Significant information be able to be obtained from this composite data by well recognized data mining techniques such as association, classification, clustering etc. By discovery hidden patterns by data mining software engineering data is made illegal. Software Engineering development has many objectives in software engineering such as Code and Design optimization, Project documentation, Development cost estimation etc. Variety of significant data mining method in each phase of software development life cycle supports in realizing these objectives proficiently and the failure rate of software is decreased. . This paper focused a new hybrid model like combination of Fuzzy Logic and knowledge management offers a significant method for developing models for software quality prediction. This research paper explains about exercise of estimate and valuation at a particular organization by developments and represents the outcomes attained with a fuzzy based classification and knowledge model for the fuzzy knowledge management predication for the quality of software engineering Approach. This result illustrate that the significance of Average Error Evaluation Efficiency observed and used in fuzzy logic is lesser than Average Error Evaluation Efficiency used in another regression multiple regression; while the value of prediction is higher value that other prediction models is used before. Thus Results demonstrate that Hybrid fuzzy knowledge management predication for the quality of software engineering can be used as alternative for predicting the Software Development and Maintenance Quality (SDMQ).

Comparison of Machine Learning Techniques for Software Quality Prediction

Software quality prediction is one the most challenging tasks in the development and maintenance of software. Machine learning (ML) is widely being incorporated for the prediction of the quality of a final product in the early development stages of the software development life cycle (SDLC). An ML prediction model uses software metrics and faulty data from previous projects to detect high-risk modules for future projects, so that the testing efforts can be targeted to those specific ‘risky' modules. Hence, ML-based predictors contribute to the detection of development anomalies early and inexpensively and ensure the timely delivery of a successful, failure-free and supreme quality software product within budget. This article has a comparison of 30 software quality prediction models (5 technique * 6 dataset) built on five ML techniques: artificial neural network (ANN); support vector machine (SVMs); Decision Tree (DTs); k-Nearest Neighbor (KNN); and Naïve Bayes Classifiers (NBC), using six datasets: CM1, KC1, KC2, PC1, JM1, and a combined one. These models exploit the predictive power of static code metrics, McCabe complexity metrics, for quality prediction. All thirty predictors are compared using a receiver operator curve (ROC), area under the curve (AUC), and accuracy as performance evaluation criteria. The results show that the ANN technique for software quality prediction is promising for accurate quality prediction irrespective of the dataset used.

A systematic literature review and comparative study of different software quality models

Every organization wants to run with profit only and it is a common trend to acquire the new technologies, methods and models to enhance the quality of software product. Quality of product is directly proportional to the value of the product and the profit of the organization. Different quality model are used by different organization foundation upon the requirements. Different thoughts of software quality attributes with various software quality models are evaluated and examined in this survey. Also researcher performed comparative study of different software quality models utilized by different organizations. Making software by using open source software (OSS) and commercial of the shelf (COTS) component is very useful for any organization. So as to quantify quality in software, quality models can be used to check the quality of existing COTS and OSS software components. The purpose of this research is to do the comparative study to check software quality prediction using COTS and OSS components. In this survey the researcher do study of more than 100 papers and identified that OSS as the development technique is more better then COTS development. Because COTS development is black box in nature, so some tradition quality metrics cannot be apply but in case of OSS as development methodology source code is available. So we can apply all software quality metrics to assess the OSS quality. But still there is no proper technique, model and tool is available by which we can evaluate the quality of OSS. In future the researcher can improve the software quality by creating some tools and models for estimation of quality characteristics according to the modified ISO-9126 quality assurance model for OSS.

Analysis and Prioritization of Design Metrics

Clones at domain level during the design phase plays an important role to identify the risk at an early stage. The authors have analyzed the Design Metrics (DM) for its contribution towards the structural properties in domain models of the software. Then prioritization of these DM leads to classification of the majorly contributing DM elements. These Design Metrics relates to classes within the class diagram at design level that is useful for the identification of cross cutting concerns, origin analysis of software, software quality prediction and maintenance at the class level granularity. The authors have empirically evaluated the results for structural model clone detection.

The Development of DT-NB Hybrid Algorithms for Classifying Some Defective Dataset Types for Software Quality Prediction

The Development of DT-NB Hybrid Algorithms for Classifying Some Defective Dataset Types for Software Quality Prediction

Svs: Prediction Framework for Software Quality Enhancement through Data Mining Techniques

Software Engineering has its origins in tackling the issue of development and maintenance of quality software. Software Quality has been defined in multiple ways but the broadest definition is that quality is the extent to which the customer is satisfied with the developed software. Data mining has the prospects of being applied to multiple domains and addressing the long standing issues faced by them. It has been successfully applied to uncover solutions to complex problems that have long confronted these domains. The proposed research is a step in the direction. It will attempt to apply existing data mining algorithms to data accumulated by software organizations in an attempt to extract useful patterns that can go a long way in addressing the issue of software quality. This work proposed Spacious Virtue Suggestion (SVS) Model for analyzing code based quality in software quality model. The first layer of this model is Extraction Layer that extracts the various attributes of software code used. After the extraction of the metrics attributes are constructed as a vector is considered as the feature vector for the second layer of the SVS Model. The second layer of the SVS model is Selection Layer which employs feature selection strategy to obtain significant metrics attributes for the software quality prediction by reducing the overlapping metrics attributes from the vector... The third layer of SVS Model is Prediction Layer which predict the good class from the training set and result shows the high accuracy in the proposed system.