Due to delay in proper treatment and diagnosis, Cancer move toward one of major character that is accountable for fatalities globally. This happened due to uncontrolled and abnormal cell growth in human body, that can get transferred to different parts of human flesh. Sequence of Ribonucleic acid (RNA) checks variation that takes place in cells and assist in analyzing transcriptome in pattern of gene expression inside RNA fluids. In an early stages of cancer schemes of Machine learning contributes in cancerous cell prediction, if concerned information is made easily accessible. Main aim of this research is in building prototypes and group various kind of cancer. Hence, we deployed many machine learning schemes such as support vector machine (SVM), random forest (RF), multilayer perceptron (MLP), & k-nearest neighbors ( KNN) for classifying samples as per their labeling. Out of Genotype-Tissue Expression (GTEX) & cancer genome Atlas (TCGA), for this thesis datasets were accumulated. On independent dataset (GTEX) machine learning techniques were tested and trained on TCGA data. Data representation thus received by applying stacked denoising autoencoders areutilized in training and testing prototypes. These models were not having very high performance; but, comparatively to others MLP exhibited better performance. Best characteristics that were chosen by applying SelectKBest, were utilized too for performance comparison. Outcome showed that K-nearest neighbor classifier provided good results, having precision of 85.12% whereas verified with self-sufficient information, with 98.4%. as accuracy of training.