Published Paper Details:

This paper presents an integrated approach to enhance software defect prediction and sentiment analysis using advanced machine learning techniques. For software defect prediction, we employ a Support Vector Machine (SVM) model to identify potential defects in software components. The SVM model is trained on historical software metrics and defect logs, incorporating feature selection and engineering to improve prediction accuracy. Additionally, we address sentiment analysis by using an optimized Logistic Regression model. Text data is preprocessed and transformed using a Count Vectorizer, with hyperparameters fine-tuned via Grid Search to enhance model performance. This dual approach demonstrates the versatility and effectiveness of machine learning in various applications, showcasing significant improvements in both software defect prediction and sentiment analysis. The combined methodologies not only improve defect prediction in software development but also provide valuable insights into textual data analysis, illustrating the broad applicability and robustness of machine learning techniques.