Survey on Feature Selection for Data Reduction

摘要: The storage capabilities and advanced in data collection has led to an information load the size of databases increases dimensions, not only rows but also columns. Data reduction (DR) plays a vital role as prepossessing techniques area knowledge discovery from huge data. Feature selection (FS) is one well known techniques, which deals with attributes original without affecting main content. Based on training used for different applications discovery, FS technique falls into supervised, unsupervised. In this paper extensive survey supervised describing searching approach, methods application areas outline comparative study covered.