Cancer Type Detection based on Gene Expression Data using Support Vector Machine

Detecting cancer types using machine learning and gene expression data is a valuable and difficult problem in bioinformatics and medical research. This procedure involves analyzing gene expression profiles to detect the type of cancer that a patient may have. Genes are crucial in the development, progression, and categorization of many cancer types. It has become challenging to find the most useful and significant gene type with a specified value in a cancer classification task. In this study, we proposed a model for detecting cancer types such as breast cancer, renal cancer, colon cancer, lung cancer, and prostate cancer based on gene type and their specified value using machine learning. In our research, the ICMR gene dataset is used. Using an integration of Minimum Redundancy Maximum Relevance and wrapper-based feature selection method, gene types are selected based on the relationship between their gene expression values and cancer types. The selected features are used for cancer classification using SVM and achieved 96% accuracy.