The second most common cause of cancer-related deaths in women, following lung cancer, is breast cancer. About 1 out of 8 women will get invasive breast cancer in their lifetime. Accurate diagnosis and classification are crucial for efficient treatment; however, traditional methods often rely on subjective histopathological examination, which can be time-consuming and prone to fluctuations. Machine learning (ML) techniques are emerging as powerful tools that can enhance breast cancer classification by leveraging high-dimensional image data and clinical features, thus improving diagnostic accuracy and reducing risks related to incorrect diagnosis. This study explores various ML techniques, particularly focusing on the Convolutional Long- Short Term Deep Neural Network (CLDNN) model integrated with Support Vector Machines (SVM). Through comparisons with other models, such as CNN+LSTM, CLDNN+ Random Forest Classifier, and CLDNN+LSTM, we found that the CLDNN+SVM combination, when integrated with advanced feature extraction techniques, achieved the best performance, resulting in an impressive accuracy of 99.30%.