Visualization of performance of binary search in worst case in personal computer using semi parametric technique

Abstract

Binary search algorithm is a well-known divide and conquer searching algorithm. In this present work, the researchers have generated the experimental data set by running java code for binary search algorithm in the worst case on Linux operating system in a personal computer (laptop) and have calculated mean values of the observed execution time i.e. one thousand observations for each data size where the data sizes are five thousand to twenty thousand with an interval of five hundred to eliminate and/or decrease the noises from the observations (for each data size). The data points i.e. Data size versus Mean execution time in nano seconds are analyzed using semi parametric approach with varying degrees, residual diagnostics have been carried out for each of these models (fits) and the results of semi parametric fits are graphically represented to visualize the performance of the Binary search algorithm in the worst case in a personal computer (laptop).