User Interface Bug Classification Model Using ML and NLP Techniques: A Comparative Performance Analysis of ML Models
Journal Title: International Journal of Experimental Research and Review - Year 2024, Vol 45, Issue 9
Abstract
Analyzing user interface (UI) bugs is an important step taken by testers and developers to assess the usability of the software product. UI bug classification helps in understanding the nature and cause of software failures. Manually classifying thousands of bugs is an inefficient and tedious job for both testers and developers. Objective of this research is to develop a classification model for the User Interface (UI) related bugs using supervised Machine Learning (ML) algorithms and Natural Language Processing (NLP) techniques. Also, to assess the effect of different sampling and feature vectorization techniques on the performance of ML algorithms. Classification is based upon ‘Summary’ feature of the bug report and utilizes six classifiers i.e., Gaussian Naïve Bayes (GNB), Multinomial Naïve Bayes (MNB), Logistic Regression (LR), Support Vector Machines (SVM), Random Forest (RF) and Gradient Boosting (GB). Dataset obtained is vectored using two vectorization techniques of NLP i.e., Bag of Words (BoW) and Term Frequency-Inverse Document Frequency (TF-IDF). ML models are trained after vectorization and data balancing. The models ' hyperparameter tuning (HT) has also been done using the grid search approach to improve their efficacy. This work provides a comparative performance analysis of ML techniques using Accuracy, Precision, Recall and F1 Score. Performance results showed that a UI bug classification model can be built by training a tuned SVM classifier using TF-IDF and SMOTE (Synthetic Minority Oversampling Techniques). SVM classifier provided the highest performance measure with Accuracy: 0.88, Precision: 0.86, Recall: 0.85 and F1: 0.85. Result also inferred that the performance of ML algorithms with TF-IDF is better than BoW in most cases. This work provides classification of bugs that are related to only the user interface. Also, the effect of two different feature extraction techniques and sampling techniques on algorithms were analyzed, adding novelty to the research work.
Authors and Affiliations
Sara Khan, Saurabh Pal
Keywords
Related Articles
- EP ID EP752538
- DOI 10.52756/ijerr.2024.v45spl.005
- Views 32
- Downloads 1
Sequential extraction for the speciation of trace heavy metals in Hoogly river sediments, India
The spatial variation of metal concentrations in the sediments is associated with the level of pollution along river stretches. Thus, the chemical speciation of seven trace heavy metals (Co, Cd, Cr, Cu, Ni, Pb and Zn) in...
Proximal Policy Optimization for Efficient Channel Allocation with Quality of Service (QoS) in Cognitive Radio Networks
A multi-variable relationship exists in Cognitive Radio Networks (CRNs) where factors such as Energy efficiency, Throughput, Delay and Signal Noise Ratio (SINR) are related. The SINR shows the quality of the signal and i...
Health Status of Grinders in Different Foundries in West Bengal
The most significant operation to produce products in manufacturing sector in West Bengal is grinding. There is always a need of high human involvement to successfully plan and execute it. The body postures of the grinde...
The Voltage stability analysis for grid-connected PV system using optimized control tested by IEEE 14 &30 bus system
In the research work presented in this paper, we present a grid-connected solar photovoltaic (PV) system, which is focused on various factors, such as the low oxide emission and high energy efficiency of the system. Sola...
The Impact of Leadership Styles, Cultural Dimensions and Values on Academic Leaders
This study investigates the cultural dimensions, values, and leadership styles of school leaders in Indian K-12 and European schools, specifically focusing on cross-cultural differences. The objective is to explore how l...