Fig. 5From: An approach to forecast human cancer by profiling microRNA expressions from NGS dataPrediction accuracy versus numbers of attributes (microRNAs). Accuracy of prediction is above 90% when atleast 24 microRNAs are used in the experiment for lung cancer samples, 26 microRNAs for hepatocellular carcinoma samples and 20 microRNAs for carcinomas of the bladder samplesBack to article page