Primary liver cancer, one of the most common malignant tumors in China, can only be roughly diagnosed through doctors’ expertise and experience at present, making it impossible to resolve the health problem that people care about. A new method that applies machine learning to the medical filed is therefore presented in this paper. The decision tree algorithm and the random forest algorithm are used to classify the data, and decision tree algorithm and improved feature selection algorithm to select important features. Comparison shows that the performance of the random forest algorithm is better than that of the decision tree algorithm, and the improved feature selection algorithm can filter out more important features on the premise of retaining accuracy.
CITATION STYLE
Liu, Y., Pan, Q., & Zhou, Z. (2018). Improved feature selection algorithm for prognosis prediction of primary liver cancer. In IFIP Advances in Information and Communication Technology (Vol. 539, pp. 422–430). Springer New York LLC. https://doi.org/10.1007/978-3-030-01313-4_45
Mendeley helps you to discover research relevant for your work.