CaPBug-A Framework for Automatic Bug Categorization and Prioritization Using NLP and Machine Learning Algorithms

Abstract

Bug reports facilitate software development teams in improving the quality of software. These reports include significant information related to problems encountered within a software, possible enhancement suggestions, and other potential issues. Bug reports are typically complex and are too detailed; hence a lot of resources are required to analyze and process them manually. Moreover, it leads to delays in the resolution of high priority bugs. Accurate and timely processing of bug reports based on their category and priority plays a significant role in improving the quality of software maintenance. Therefore, an automated process of categorization and prioritization of bug reports is needed to address the aforementioned issues. Automated categorization and prioritization of bug reports have been explored recently by many researchers; however, limited progress has been made in this regard. In this research, we present a novel framework, titled CaPBug, for automated categorization and prioritization of bug reports. The framework is implemented using Natural Language Processing (NLP) and supervised machine learning algorithms. A baseline corpus is built with six categories and five prioritization levels by analyzing more than 2000 bug reports of Mozilla and Eclipse repository. Four classification algorithms i.e., Naive Bayes, Random Forest, Decision Tree, and Logistic Regression have been used to categorize and prioritize bug reports. We demonstrate that the CaPBug framework achieved an accuracy of 88.78% by using a Random Forest classifier with a textual feature for predicting the category. Similarly, using the CaPBug framework, an accuracy of 90.43% was achieved in predicting the priority of bug reports. Synthetic Minority Over-Sampling Technique (SMOTE) has been applied to address the class imbalance issue in priority classes.

RESMOTE: A New Approach for Class Imbalance Problem

“…collecting details about the neighbor node would be a biggest challenge. The KNN [12] and KMeans [11] methodology differences has taken as a essential in classifying specimens.Now,consider K-means-based methods to solve imbalance problems.FSCL [25]:It is a kmeans [29] based algorithm to improve the framework of KMeans [13]. rPCL:To improve fSCL, rPCL [20] has been used.…”

“…NMOTe [22]:It is used to produce invariant results. R_SMOTE [11]: Improved SMOTE technique to avoid poor performance for severity prediction.Under-sampling: The under-sampling technique [6] is technique to solve the class imbalance problems. The methodology concentrates on the majority class [7] to decrease the number of specimens by removing them.…”

The Use of AI in Software Engineering: Synthetic Knowledge Synthesis of Recent Research Literature

“…Natural language processing (NLP) in software engineering Natural language processing in software development NLP technology can drastically improve software development tasks [39]. It can support bug categorization [40], development of more secure software [41], program decomposition [42], classifying commitments [43], programming and coding [44], writing coherent and factually correct readmes [45], model-driven engineering [46], deployment of design patterns [47] and traceability management [48]-Natural language processing in software requirements engineering NLP can support human-performed linguistic analysis in requirements engineering [49,50], such as identifying domain concepts [51][52][53][54], establishing traceability links [55], requirement classification [56,57], handling ambiguity [58,59], preference extraction form scenarios [60], classification of nonfunctional requirements [61], standardization of requirements in agile approaches [62] and requirement elicitation [63].…”

An Extended Survey Concerning the Significance of Artificial Intelligence and Machine Learning Techniques for Bug Triage and Management

“…A multitude of studies consider this metric to validate the model [102], [103], [104], [105], [106], [108]. However, some of the mentioned studies also use other metrics for parallel comparisons, and/or validations, precisely because the accuracy itself presents some downfalls.…”