Breast cancer is one of the most prevalent cancers in women worldwide; thus, it's important to diagnose the disease quickly and accurately so patients have the best chance of survival. Using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset, our experiment aims to see how well three machine learning algorithms—very LR is a statistical model used for binary classification, naïve Bayes (NB) probabilistic classifier based on Bayes theorem, assuming independence between predictors, and decision tree (DT) A flowchart-like tree structure where internal nodes represent decisions based on features, and leaves represent outcomes. do at classifying breast cancer patients. 30 numerical traits extracted from FNA images describe each of the 569 cases in Cottontail-DB. At 97.62% precision, 96.47% accuracy, 96.47% F1 score, and 99.74% ROC AUC, LR outperforms NB and DT. NB, on the other hand worked relatively better (mean RA of 99%) in general but particularly gave worst service for ROC AUC(98.76%), an improvement from DT which had low accuracy and few performance measures so score comparison could be done fairly easily. These findings reveal the robustness and reliability of LR for BC screening. The use of appropriate machine-learning models is vital for increasing the precision of diagnosis and aiding in clinical decision-making, underline the investigators.