Breast carcinoma is one kind of cancer that starts in cells of breast. It is one of the extremely scarce in human life which found significantly in women due to high sickness and mortality as per changing of life style and hormonal factors. Precise forecasting and identification are essential for various attentions of the medical notifications. An assortment of technological tools helps medical practitioners for analyzing, predicting, and identifying breast carcinoma problems more effectively. In this innovative work, we proposed a model for predicting breast carcinoma using ensemble machine learning models. To begin with, we have taken Wisconsin Breast Cancer Data that drawn as of UCI ML repository that utilized to assess how well various machine learning methods work. As images are also extracted like text data, primarily one screening tool is required, known as mammography, which can detect the above disease before time. Additionally, we developed the ensemble model using SVM, RF, KNN, logistic regression, and transfer learning which may worn for enhancing the quality of images due to its speed, efficiency, and multitasking learning mechanism. The corresponding experiments have demonstrated and compared with all the traditional models. According to evaluation assessments, the ensemble approach performs exceptionally well in the assessing outcomes. Finally from the investigation result, the performance of transfer learning+ SVM+ RF is better and got accuracy 98.58% after comparison of other methods. This work affects on diseases derived from machine learning and biomedical engineering.