AI models are utilized for software defect prediction, and are there emerging models gaining traction?
Title: Trustworthy and interpretability of artificial intelligence (AI) in the Software Defect Prediction.
Objective:
Evaluate the effectiveness of AI models in recent software defect prediction and Investigate how AI contributes to defect reduction and enhances software quality.
To measure the interpretability in AI specific to defect prediction and explore methods for improving interpretability to increase the trustworthiness of AI-based predictions.
Identify and analyse the primary challenges associated with maintaining trustworthiness and interpretability in AI-based defect prediction. Examine ethical considerations surrounding the use of AI in defect prediction and propose mitigation strategies.
Research Questions:
Which AI models are utilized for software defect prediction, and are there emerging models gaining traction?
How is the interpretability of AI algorithms be quantitatively measured, and what strategies can enhance interpretability in the context of AI defect prediction?
What are the challenges of trustworthiness and transparency in AI-based defect prediction? Is the current state of defect prediction mature enough, and if not, what areas require further development?
