One essential and practical technique to improve the quality and dependability of software is to anticipate software issues. Improving project management entails identifying release delays early on and implementing cost-effective remedies to improve program quality. One approach is to estimate with information which areas of a complex software system are most likely to contain a substantial number of errors in future versions. However, it is challenging to develop models that can accurately detect errors. The primary purpose of this paper is to explore the efficiency of predictive analysis in software development systems in connection to two types of software issues: criticality and severity. The machine learning algorithm used in this paper was written in Python. This work employs statistical approaches, modeling, machine learning, data mining, and artificial intelligence. Predictive models can be used to optimize the distribution of research resources. During machine learning model training, the seriousness and urgency of a problem are assessed using two approaches: Random Forest (RF) Classifier and Support Vector Machine (SVM). The paper findings demonstrate that, with an accuracy rate of 0.87, The RF Priority Model provides a comprehensive understanding of the model's performance across various priority levels. This inquiry use data mining techniques to detect flaws in the present software configuration. Allowing developers to generate software will improve software quality while lowering development and maintenance expenses.
