Globally, breast cancer has been identified as one of the deadliest disease and top causes of cancer related mortality in females. 16% of malignant lesions that are diagnosed in the world are linked to consequence of breast cancer. Because of this, it is of the utmost importance to make a diagnosis of malignant tumours at the earliest possible stage in order to give oneself the best possible chance of surviving them. Thus, an accurate and timely diagnosis of the condition ensures the patient's long survival. Breast cancer diagnosis depends on the ability to recognise benign and malignant tumours at the appropriate time. Traditional approaches in the diagnosis of breast cancers have several drawbacks including human errors related discrepancies, inaccurate diagnosis, and time factor. Recently, machine learning algorithms together with different hyperparameter tuning optimization techniques were proved as a viable option that support the early detection of cancer.