AI-Driven Bug Hunting: Leveraging Machine Learning for Predictive Defect Detection in AR/VR

Authors: Santosh Kumar Jawalkar

DOI: https://doi.org/10.5281/zenodo.14945225

Short DOI: https://doi.org/g86n75

Country: USA

Full-text Research PDF File:   View   |   Download

Abstract: The emergence of AR (Augmented Reality) and VR (Virtual Reality) technologies is reshaping several industries, yet these advancements entail considerable hurdles, especially regarding defect detection, performance bottlenecks, and real-time debugging. In the challenge of ensuring the performance of the AR/VR application, traditional approaches like manual testing and performance profiling are often unable to cope with the complexity and dynamicity of AR/VR systems. Our work focuses on utilizing Artificial Intelligence (AI) and Machine Learning (ML) models as automated means for defect detection in AR/VR settings to enhance system reliability, user experience, and overall system capability. Various models have been implemented, such as Random Forest, XGboost, LSTMs, Autoencoders, Isolation Forest, BERT, and Latent Dirichlet allocation (LDA) to tackle several different challenges in defect detection from predicting system failures, detecting anomalies, and even classifying bug reports. We measure the accuracy, latency, false positive and false negative rates of the models and contrast their capability with the existing debugging approach. This clearly indicates the effectiveness of integrating AI solutions compared to conventional methods by exhibiting a considerable amount of reduction in defect detection time, improving overall accuracy, reducing manual efforts, and real-time analysis. This study demonstrates the support that AI-driven systems can offer in improving the process of AR/VR application development, stabilize its growth and create a better end-user experience that leads to scalable and efficient AR/VR ecosystems in the long run.

Keywords: Augmented Reality (AR), Virtual Reality (VR), Artificial Intelligence (AI), Machine Learning (ML),, Defect Detection, Predictive Analytics, Anomaly Detection, Performance Bottleneck Prediction, Random Forest, XGBoost, LSTM (Long Short-Term Memory), Autoencoders, Isolation Forest, BERT, Natural Language Processing (NLP), Latent Dirichlet Allocation (LDA), Real-Time Monitoring, Scalable Solutions, User Experience, Software Reliability.

Paper Id: 232178

Published On: 2023-09-12

Published In: Volume 11, Issue 5, September-October 2023

Share this