Advancements in Predictive Models for Software Defects: A Comprehensive Exploration
Authors:-Professor Mohamed Abdul Jailani, Kallal Bhakta
Abstract-This venture centers on foreseeing program bugs utilizing a dataset from the College of Geneva, Switzerland, which incorporates data from different program frameworks like Overshadow JDT Center, Overshadow PDE UI, and Lucene. By analyzing program properties such as lines of code, strategies, and traits, the point is to anticipate the number of bugs in progress. This foreknowledge makes a difference in proactive imperfection administration and chance relief. The venture envelops intensive information investigation, preprocessing, and visualization, taken after by progressed exploratory information investigation (EDA) utilizing machine learning and dimensionality lessening procedures. It addresses challenges like hyperparameter tuning and lesson awkwardness, endeavoring to classify computer program information by bug seriousness, from no bugs to different bugs. The extreme objective is to make strides in program support and streamline discharge forms.
