An analysis on new hybrid parameter selection model performance over big data set

摘要: Abstract Parameter selection or attribute is one of the crucial tasks in data analysis process. Incorrect important might generate imprecise event for a wrong decision. It an advantage if decision-maker could select and apply best model that helps identifying best-optimized set — decision Recently, many scientists from various application areas are attracted to investigate analyze advantages disadvantages big data. One issues is, analyzing large volumes variety environment very challenging when there lack suitable no appropriate be implemented used as guideline. Hence, this paper proposes alternative parameterization able most optimized without requiring high cost learn, use, maintain. The based on two integrated models combined with correlation-based feature selection, best-first search algorithm, soft set, rough theories which were compliments each other parameter method. Experimental have shown proposed has significantly