The impact of feature reduction techniques on defect prediction models

Kondo, Masanari; Bezemer, Cor Paul; Kamei, Yasutaka; Hassan, Ahmed E.; Mizuno, Osamu

doi:10.1007/s10664-018-9679-5

Cited by 81 publications

(71 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Clearly, there was no significant difference in the performance of the FS methods, as their respective performance and effect varies from dataset to dataset and the choice of classification algorithm. This research outcome is related to the findings from Xu et al [29], Kondo et al [31] and Muthukumaran et al [30]. Although on average, FSS methods proved to be better than FFR methods.…”

Section: Resultssupporting

confidence: 69%

“…Recent studies have compared the impact of FS methods on the performance of SDP [26][27][28][29][30][31][32]. Some studies conclude that some of the FS methods are better than others [27,28,30,31], while some studies claimed that there is no significant difference between the performances of FS methods in SDP [26,29,32]. This contradiction and inconsistency in results by existing studies may be due to the choice of search mechanism used in FS methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Performance Analysis of Feature Selection Methods in Software Defect Prediction: A Search Method Approach

et al. 2019

View full text Add to dashboard Cite

Software Defect Prediction (SDP) models are built using software metrics derived from software systems. The quality of SDP models depends largely on the quality of software metrics (dataset) used to build the SDP models. High dimensionality is one of the data quality problems that affect the performance of SDP models. Feature selection (FS) is a proven method for addressing the dimensionality problem. However, the choice of FS method for SDP is still a problem, as most of the empirical studies on FS methods for SDP produce contradictory and inconsistent quality outcomes. Those FS methods behave differently due to different underlining computational characteristics. This could be due to the choices of search methods used in FS because the impact of FS depends on the choice of search method. It is hence imperative to comparatively analyze the FS methods performance based on different search methods in SDP. In this paper, four filter feature ranking (FFR) and fourteen filter feature subset selection (FSS) methods were evaluated using four different classifiers over five software defect datasets obtained from the National Aeronautics and Space Administration (NASA) repository. The experimental analysis showed that the application of FS improves the predictive performance of classifiers and the performance of FS methods can vary across datasets and classifiers. In the FFR methods, Information Gain demonstrated the greatest improvements in the performance of the prediction models. In FSS methods, Consistency Feature Subset Selection based on Best First Search had the best influence on the prediction models. However, prediction models based on FFR proved to be more stable than those based on FSS methods. Hence, we conclude that FS methods improve the performance of SDP models, and that there is no single best FS method, as their performance varied according to datasets and the choice of the prediction model. However, we recommend the use of FFR methods as the prediction models based on FFR are more stable in terms of predictive performance.

show abstract

Section: Resultssupporting

confidence: 69%

Section: Introductionmentioning

confidence: 99%

Performance Analysis of Feature Selection Methods in Software Defect Prediction: A Search Method Approach

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Clearly, there was no significant difference in the performance of the feature subset methods, as their respective performance and effect varies from dataset to dataset and the choice of classification algorithm. This research outcome is similar to the findings from [60,61,62].…”

Section: With Feature Subset Selectionsupporting

confidence: 87%

Software Defect Prediction Based on Feature Subset Selection and Ensemble Classification

Saifan¹,

Abu-wardih²

2020

ECTI-CIT

View full text Add to dashboard Cite

Two primary issues have emerged in the machine learning and data mining community: how to deal with imbalanced data and how to choose appropriate features. These are of particular concern in the software engineering domain, and more specifically the field of software defect prediction. This research highlights a procedure which includes a feature selection technique to single out relevant attributes, and an ensemble technique to handle the class-imbalance issue. In order to determine the advantages of feature selection and ensemble methods we look at two potential scenarios: (1) Ensemble models constructed from the original datasets, without feature selection; (2) Ensemble models constructed from the reduced datasets after feature selection has been applied. Four feature selection techniques are employed: Principal Component Analysis (PCA), Pearson’s correlation, Greedy Stepwise Forward selection, and Information Gain (IG). The aim of this research is to assess the effectiveness of feature selection techniques using ensemble techniques. Five datasets, obtained from the PROMISE software depository, are analyzed; tentative results indicate that ensemble methods can improve the model's performance without the use of feature selection techniques. PCA feature selection and bagging based on K-NN perform better than both bagging based on SVM and boosting based on K-NN and SVM, and feature selection techniques including Pearson’s correlation, Greedy stepwise, and IG weaken the ensemble models’ performance.

show abstract

“…30 www.mediawiki.org (accedido 1/2020) 31 es.wikipedia.org (accedido 1/2020) 32 wikimediafoundation.org (accedido 1/2020) 33 www.mediawiki.org/wiki/MediaWiki (accedido 1/2020) 34 Según Siteviews: tools.wmflabs.org/siteviews (accedido 1/2020) 35 Según Alexa: www.alexa.com/siteinfo/wikipedia.org (accedido 1/2020). 36 mariadb.org (accedido 1/2020) 37 GNU General Public License, disponible online en www.gnu.org/licenses/gpl-3.0.html (accedido 1/2020) MediaWiki es desarrollado siguiendo un modelo de integración continua, donde los cambios en el código fuente se despliegan directamente en los sitios de Wikimedia Foundation, tales como Wikipedia, de forma regular.…”

Section: Figura 43: Captura De Pantalla De La Herramienta Para Clasiunclassified

“…Recientemente Kondo [36] ha explorado el modo en que se pueden seleccionar o combinar las métricas con las que se construyen modelos (técnicas de feature selection y feature reduction). La necesidad de reducir la cantidad de métricas aparece cuando la cantidad de información (muestra) es pequeña, algo que sucede en proyectos nuevos y atenta contra los resultados que pueden proveer los modelos.…”

Section: Figura 68: Gráfica De La Matriz De Confusión Obtenida a Parunclassified

Predicción de defectos en un lenguaje dinámicamente tipado usando métricas estáticas y de cambio

Gullino¹

View full text Add to dashboard Cite

La predicción de defectos es un tema importante en la Ingeniería de Software. Puede utilizarse para evaluar la calidad del producto final, estimando si se cumplen los estándares de calidad contractuales o los impuestos para la aceptación del producto por parte del cliente. También puede facilitar la previsión de los recursos para pruebas o verificación formal. El presente trabajo selecciona un conjunto de métricas estáticas (o de producto) y de cambio (o de proceso) con el objetivo de predecir los defectos de versiones sucesivas. Como método de estimación se utiliza la regresión logística. Se presenta un caso de estudio sobre el proyecto de código abierto MediaWiki, que soporta a Wikipedia. Este producto cuenta con 1000 clases y 365 KLOC. Se han obtenido las métricas estáticas y de cambio durante el período de un año con distintos cortes temporales. Se construyeron herramientas ad hoc para calcular las métricas definidas por autores como Chidamber & Kemerer, Robiolo y Moser. Se busca estudiar la viabilidad de aplicar estas métricas en un lenguaje de programación orientado a objetos diferente, ya que las investigaciones de referencia han trabajado históricamente con lenguajes de tipado estático (principalmente C++ y Java) mientras que MediaWiki se encuentra desarrollado en PHP, que es un lenguaje orientado a objetos dinámicamente tipado, muy utilizado en el desarrollo web de pequeña y mediana escala. A estos efectos, también se proponen métricas nuevas, que buscan medir estos aspectos de tipado del lenguaje y sus efectos en el producto y la calidad del mismo. Para recolectar las métricas de cambio del código fuente se han clasificado manualmente alrededor de 2800 commits de 230 desarrolladores, realizados al repositorio de versiones del proyecto durante el año 2014. Esta clasificación tiene por objeto identificar qué modificaciones al código fuente se pueden imputar a correcciones de defectos (Fault Repairing) y cuáles a introducción de nueva funcionalidad (Feature Introduction), siguiendo el trabajo de Hassan. Esta clasificación permite definir métricas con las cuales construir modelos de predicción de defectos. Se comprueba que la cantidad y tamaño de los cambios de tipo “Feature Introduction” realizados a una clase son los mejores predictores de los defectos futuros de la misma. Además, se comprueba que utilizando métricas de cambio es posible obtener mejores resultados en la predicción que con métricas estáticas, pero las primeras representan un mayor esfuerzo de medición que las segundas.

show abstract

The impact of feature reduction techniques on defect prediction models

Cited by 81 publications

References 59 publications

Performance Analysis of Feature Selection Methods in Software Defect Prediction: A Search Method Approach

Performance Analysis of Feature Selection Methods in Software Defect Prediction: A Search Method Approach

Software Defect Prediction Based on Feature Subset Selection and Ensemble Classification

Predicción de defectos en un lenguaje dinámicamente tipado usando métricas estáticas y de cambio

Contact Info

Product

Resources

About