Cancer Classification using Gene Expression Dataset to Detect Cancer

Cancer classification plays a very important role in personalized medicine and targeted therapy. To classify the kinds of cancer based on the gene expression data, three different kinds of algorithms are applied in this paper: the greedy method, the Recursion Tree Method, and Hashing. The selection of these algorithms forms a large combination to address the complexity of the gene expression datasets in an attempt to increase the accuracy of cancer classification and minimize the computational time. The study evaluates the performance of each of the algorithms and also show how well they do in cancer classification through detailed experimentation and review. The Greedy Algorithm and Feature Hashing enable accuracy of 88% and 85%, respectively. Feature Hashing reduces the computational complexity significantly. Such findings bring out the trade-offs between accuracy and computational efficiency, giving insights that might be of importance for cancer detection and treatment methods.