Findings from Northeastern University Has Provided New Data on Technology (A Hybrid Feature-selection Method Based On Mrmr and Binary Differential Evolution for Gene Selection).

A new report from Northeastern University in Shenyang, China discusses the importance of selecting critical features from microarray data as biomarkers for disease diagnosis and drug development. The report proposes a hybrid feature-selection algorithm that combines the Max Relevance and Min Redundancy (mRMR) method with differential evolution. This algorithm improves the quantization functions of mRMR to accommodate the continuous nature of microarray data attributes and utilizes an enhanced differential evolution algorithm to further filter the features. The research concludes that this hybrid algorithm improves the performance and efficiency of feature selection for microarray data analysis. [Extracted from the article]