A system has less efficient decision-making rules that are effectively understandable and manageable in industries such as banking, medical, financial, etc. A decision tree (DT) is a method of supervised machine learning employed to produce a clear and interpretable framework of rules for efficient decision-making. It occasionally generates a few unnecessary and redundant rules, reducing its readability. This paper introduces an expert system named the Transparent Rule Extraction System (TRES), designed to address existing challenges in the early diagnosis of breast cancer (BC). This work significantly reduces the number of rules without compromising accuracy. To eliminate redundant decision rules, the proposed approach integrates a sequential hill climbing strategy with a customized heuristic function. The proposed TRES establishes a coherent and comprehensible set of rules for the early identification of BC. The proposed model achieved a satisfactory accuracy of 96.47% on the benchmark BC dataset.