An Improved Quantum Software Challenges Classification Approach using Transfer Learning and Explainable AI

2025-09-25

AI Summary

Quantum software engineers frequently discuss quantum software engineering (QSE) challenges on platforms like Stack Overflow; however, existing quantum-related tags emphasize low-level technical implementation and lack a practice-oriented taxonomy for categorizing QSE challenges. Method: We propose a novel, interpretable approach integrating transfer learning and explainable AI (XAI): fine-tuning BERT, DistilBERT, and RoBERTa transformer models, augmented with SHAP-based interpretability analysis; complemented by manual annotation and validation of 2,829 real-world questions via content analysis and grounded theory. Contribution/Results: Our resulting taxonomy achieves a mean accuracy of 95%, outperforming FNN, CNN, and LSTM baselines by 6%. It is the first automated classification framework that simultaneously delivers high accuracy, model interpretability, and direct relevance to QSE practice challenges—thereby significantly enhancing knowledge organization and retrieval efficiency in quantum development communities.

Technology Category

• Machine Learning: Quantum Machine Learning
• Knowledge Representation and Reasoning: Qualitative Reasoning
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Community question answering
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Quantum Software Engineering (QSE) is a research area practiced by tech firms. Quantum developers face challenges in optimizing quantum computing and QSE concepts. They use Stack Overflow (SO) to discuss challenges and label posts with specialized quantum tags, which often refer to technical aspects rather than developer posts. Categorizing questions based on quantum concepts can help identify frequent QSE challenges. We conducted studies to classify questions into various challenges. We extracted 2829 questions from Q&A platforms using quantum-related tags. Posts were analyzed to identify frequent challenges and develop a novel grounded theory. Challenges include Tooling, Theoretical, Learning, Conceptual, Errors, and API Usage. Through content analysis and grounded theory, discussions were annotated with common challenges to develop a ground truth dataset. ChatGPT validated human annotations and resolved disagreements. Fine-tuned transformer algorithms, including BERT, DistilBERT, and RoBERTa, classified discussions into common challenges. We achieved an average accuracy of 95% with BERT DistilBERT, compared to fine-tuned Deep and Machine Learning (D&ML) classifiers, including Feedforward Neural Networks (FNN), Convolutional Neural Networks (CNN), and Long Short-Term Memory networks (LSTM), which achieved accuracies of 89%, 86%, and 84%, respectively. The Transformer-based approach outperforms the D&ML-based approach with a 6% increase in accuracy by processing actual discussions, i.e., without data augmentation. We applied SHAP (SHapley Additive exPlanations) for model interpretability, revealing how linguistic features drive predictions and enhancing transparency in classification. These findings can help quantum vendors and forums better organize discussions for improved access and readability. However,empirical evaluation studies with actual developers and vendors are needed.

Problem

• Classifying quantum software challenges from developer discussions on Q&A platforms
• Developing accurate classification models using transfer learning and transformer algorithms
• Enhancing model interpretability with explainable AI techniques for transparency

Innovation

• Fine-tuned transformer models classify quantum software challenges
• SHAP provides model interpretability for classification transparency
• ChatGPT validates human annotations to resolve disagreements