Pneumonia mainly caused by Streptococcus pneumoniae is a serious lung infection and one of the major causes of morbidity and mortality worldwide. Their global burden is so massive considering one of three deaths reported by the World Health Organization (WHO) in India which underlines an urgent need for accurate and effective diagnostic method for pneumonia. Chest X-ray interpretation for diagnosis is commonly undertaken by expert radiologists but is difficult in rural places. In this research, we propose a deep learning-based method for pneumonia detection based on X-ray chest images. The accuracy, precision, F1-score, and AUC in 30 epochs of the proposed model that combines VGG- 19 with a Random Forest classifier as part of the CNN is 90.03%, 93%, 91% and 0.95 respectively, thereby and thus proves its utility for pneumonia classification. The model has proven its ability in extracting features and in detection toward smaller X-ray dataset and in real-life healthcare applications. The novelty of this work is through the combination of the hybrid technique utilizes in which to better perform pneumonia detection through accuracy and efficiently than other typical techniques utilizing a hybrid paradigm to address the challenge pneumonia detection more reliably in different healthcare settings.