Feature selection for large-scale data sets in GrC

Granular computing, as an emerging computational and mathematical theory which describes and processes uncertain, vague, incomplete, and mass information, has been successfully used in knowledge discovery. At present, granular computing faces the challenges of consuming a huge amount of computational time and memory space in dealing with large-scale and complicated data sets. Feature selection, a common technique for data preprocessing in many areas such as pattern recognition, machine learning and data mining, is of great importance. This paper focuses on efficient feature selection algorithms for large-scale data sets and dynamic data sets in granular computing.