• Mr. Yerranagu Sai Kiran
• Mr. P. Naveen Sundar Kumar

Various stages of the software development process call for the activity of software quality estimate. It could be utilized for benchmarking and planning the project's quality assurance procedures. Multiple Criteria Linear Programming and Multiple Criteria Quadratic Programming were the two techniques employed in early studies to determine the program quality. Additionally, experiments using C5.0, SVM, and Neutral networks for quality estimation were conducted. These studies have a paltry degree of accuracy. In this study, we used pertinent information from a huge dataset to increase estimation accuracy. To achieve improved accuracy, we used a feature selection method and a correlation matrix. We have also tested more contemporary techniques that have been effective for various prediction challenges. Algorithms for machine learning like XGBoost, Random Forest, and Decision Tree To forecast software quality and identify the relationship between quality and development parameters, data is processed using logistic regression and naive bayes. These algorithms produce low recall and precision ratings. We employ the cat boost algorithm and the SMOTE approach to get around this. This suggested technique outperforms the current approach while also scoring well in terms of recall and precision.

Estimation, Machine Learning, Software Quality, Extreme Gradient Decent, Boosting

"Predicting Software Quality Using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 7, page no.j89-j92, July-2023, Available :http://www.jetir.org/papers/JETIR2307913.pdf

"Predicting Software Quality Using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 7, page no. ppj89-j92, July-2023, Available at : http://www.jetir.org/papers/JETIR2307913.pdf