An empirical study of double-bridge search move on subset feature selection search of bees algorithm

The application of bees algorithm in wrapper feature selection (bafs) has been implemented but yet too far from perfect and has few weaknesses. the algorithm performs combination of exploitative neighbourhoods and random explorative search. this creates a heavy computational time,and in the same time could affect the overall accuracy subset selection. to rectify this issue,a double-bridge move proposed and benchmark dataset have been used to determine the performance of the proposed method. the obtained results from the experiment confirmed that the proposed extension of the search neighbourhood have provided better accuracy with suitable time than the original bafs.