Published Paper Details:

Faulty software detection is a critical task in software development because it can highly impact the quality and reliability of software systems. In this paper, we propose a novel approach for faulty software detection using machine learning and the XGBoost algorithm. Our approach utilizes a dataset of Python code with control structures of basic such as if, else, and do while loops, and applies the XGBoost algorithm to diagnose faults in the code. We evaluate the performance of our approach using metrics such as accuracy, precision, recall, F1-score, and ROC AUC score, and compare it with existing research. Our results show that the XGBoost algorithm is highly effective in diagnosing faults in Python code, and our approach achieves state-of-the-art results on many machine learning challenges. We also provide insights on cache access patterns, data compression and sharding to build a scalable tree boosting system. By combining these insights, our approach scales beyond billions of examples using far fewer resources than existing systems.