Sample Based Curve Fitting Computation on the Performance of Quicksort in Personal Computer

Abstract

In this paper we have used curve fitting technique for analyzing the classical Quicksort algorithm and its performance in worst case on personal computer. The proposed generic model can be viewed as: Time ~ f (Number of Elements). We fit the data points (Time vs Number of elements) in twenty one different models from various types of fit such as Polynomial, Exponential, Power, Gaussian and Fourier. This analysis leads us to identify the best model amongst these models. We found that a model of Fourier series is the best fit.