BFSp: A feature selection method for bug severity classification

Sharmin, Sadia; Aktar, Farzana; Ali, Amin Ahsan; Khan, Muhammad Asif; Shoyaib, Mohammad

doi:10.1109/r10-htc.2017.8289066

Cited by 12 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It was observed that the similarity measure improved the severity prediction and fixer recommendation process. Sharmin et al [34] drafted a bug feature selection (BFS) technique by using Pareto optimality to classify the bugs into Blocker, Trivial, Critical, Minor, and Major classes by searing informative features. The performance was evaluated on three open source projects, i.e., Eclipse, GCC, and Mozilla, and results showed that the BFS technique performed better than existing techniques [33].…”

Section: Related Workmentioning

confidence: 99%

A Novel Deep-Learning-Based Bug Severity Classification Technique Using Convolutional Neural Networks and Random Forest with Boosting

Kukkar

Mohana

Nayyar

et al. 2019

Sensors

View full text Add to dashboard Cite

The accurate severity classification of a bug report is an important aspect of bug fixing. The bug reports are submitted into the bug tracking system with high speed, and owing to this, bug repository size has been increasing at an enormous rate. This increased bug repository size introduces biases in the bug triage process. Therefore, it is necessary to classify the severity of a bug report to balance the bug triaging process. Previously, many machine learning models were proposed for automation of bug severity classification. The accuracy of these models is not up to the mark because they do not extract the important feature patterns for learning the classifier. This paper proposes a novel deep learning model for multiclass severity classification called Bug Severity classification to address these challenges by using a Convolutional Neural Network and Random forest with Boosting (BCR). This model directly learns the latent and highly representative features. Initially, the natural language techniques preprocess the bug report text, and then n-gram is used to extract the features. Further, the Convolutional Neural Network extracts the important feature patterns of respective severity classes. Lastly, the random forest with boosting classifies the multiple bug severity classes. The average accuracy of the proposed model is 96.34% on multiclass severity of five open source projects. The average F-measures of the proposed BCR and the existing approach were 96.43% and 84.24%, respectively, on binary class severity classification. The results prove that the proposed BCR approach enhances the performance of bug severity classification over the state-of-the-art techniques.

show abstract

Section: Related Workmentioning

confidence: 99%

A Novel Deep-Learning-Based Bug Severity Classification Technique Using Convolutional Neural Networks and Random Forest with Boosting

Kukkar

Mohana

Nayyar

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…Many research works, as described in [21][22][23][24][25] employed different feature selection methods to minimize the number of informative features set and to enhance the accuracy of the classifier. All the works mentioned above paid particular attention to the effectiveness of feature selection techniques on the accuracy of classifying the severity of bug reports.…”

Section: Related Workmentioning

confidence: 99%

Machine Learning Approaches for Predicting the Severity Level of Software Bug Reports in Closed Source Projects

Baarah¹,

Aloqaily²,

Salah³

et al. 2019

IJACSA

View full text Add to dashboard Cite

In Software Development Life Cycle, fixing defect bugs is one of the essential activities of the software maintenance phase. Bug severity indicates how major or minor the bug impacts on the execution of the system and how rapidly the developer should fix it. Triaging a vast amount of new bugs submitted to the software bug repositories is a cumbersome and time-consuming process. Manual triage might lead to a mistake in assigning the appropriate severity level for each bug. As a consequence, a delay for fixing severe software bugs will take place. However, the whole process of assigning the severity level for bug reports should be automated. In this paper, we aim to build prediction models that will be utilized to determine the class of the severity (severe or non-severe) of the reported bug. To validate our approach, we have constructed a dataset from historical bug reports stored in JIRA bug tracking system. These bug reports are related to different closed-source projects developed by INTIX Company located in Amman, Jordan. We compare eight popular machine learning algorithms, namely Naive Bayes, Naive Bayes Multinomial, Support Vector Machine, Decision Tree (J48), Random Forest, Logistic Model Trees, Decision Rules (JRip) and K-Nearest Neighbor in terms of accuracy, F-measure and Area Under the Curve (AUC). According to the experimental results, a Decision Tree algorithm called Logistic Model Trees achieved better performance compared to other machine learning algorithms in terms of Accuracy, AUC and F-measure with values of 86.31, 0.90 and 0.91, respectively.

show abstract

“…create noise or be redundant. FS methods are used to remove such irrelevant and redundant features and can be divided into three broad categories namely Wrapper [14,18], Embedded [20], and Filter methods [13,15,16]. Among these, filter methods do not depend on a classifier to select a feature.…”

mentioning

confidence: 99%

Discretization and Feature Selection Based on Bias Corrected Mutual Information Considering High-Order Dependencies

Roy

Sharmin

Ali

et al. 2020

Advances in Knowledge Discovery and Data Mining

Self Cite

View full text Add to dashboard Cite

Mutual Information (MI) based feature selection methods are popular due to their ability to capture the nonlinear relationship among variables. However, existing works rarely address the error (bias) that occurs due to the use of finite samples during the estimation of MI. To the best of our knowledge, none of the existing methods address the bias issue for the high-order interaction term which is essential for better approximation of joint MI. In this paper, we first calculate the amount of bias of this term. Moreover, to select features using χ 2 based search, we also show that this term follows χ 2 distribution. Based on these two theoretical results, we propose Discretization and feature Selection based on bias corrected Mutual information (DSbM). DSbM is extended by adding simultaneous forward selection and backward elimination (DSbM fb ). We demonstrate the superiority of DSbM over four state-of-the-art methods in terms of accuracy and the number of selected features on twenty benchmark datasets. Experimental results also demonstrate that DSbM outperforms the existing methods in terms of accuracy, Pareto Optimality and Friedman test. We also observe that compared to DSbM, in some dataset DSbM fb selects fewer features and increases accuracy.

show abstract

BFSp: A feature selection method for bug severity classification

Cited by 12 publications

References 24 publications

A Novel Deep-Learning-Based Bug Severity Classification Technique Using Convolutional Neural Networks and Random Forest with Boosting

A Novel Deep-Learning-Based Bug Severity Classification Technique Using Convolutional Neural Networks and Random Forest with Boosting

Machine Learning Approaches for Predicting the Severity Level of Software Bug Reports in Closed Source Projects

Discretization and Feature Selection Based on Bias Corrected Mutual Information Considering High-Order Dependencies

Contact Info

Product

Resources

About